def metamove(paramsdict, partids, partstack, outputdir, procid, myid, main_node, nproc): # Reads from paramsdict["stack"] particles partids set parameters in partstack # and do refinement as specified in paramsdict # # Will create outputdir # Will write to outputdir output parameters: params-chunk0.txt and params-chunk1.txt if(myid == main_node): # Create output directory log = Logger(BaseLogger_Files()) log.prefix = os.path.join(outputdir) cmd = "mkdir "+log.prefix cmdexecute(cmd) log.prefix += "/" else: log = None mpi_barrier(MPI_COMM_WORLD) ali3d_options.delta = paramsdict["delta"] ali3d_options.center = paramsdict["center"] ali3d_options.ts = paramsdict["ts"] ali3d_options.xr = paramsdict["xr"] ali3d_options.fl = paramsdict["currentres"] ali3d_options.aa = paramsdict["aa"] ali3d_options.maxit = paramsdict["maxit"] ali3d_options.mask3D = paramsdict["mask3D"] projdata = getindexdata(paramsdict["stack"], partids, partstack, myid, nproc) if(paramsdict["delpreviousmax"]): for i in xrange(len(projdata)): try: projdata[i].del_attr("previousmax") except: pass ali3d_options.ou = paramsdict["radius"] # This is changed in ali3d_base, but the shrank value is needed in vol recons, fixt it! if(myid == main_node): line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"METAMOVE parameters") spaces = " " for q in paramsdict: print(" => ",q+spaces[len(q):],": ",paramsdict[q]) print(" => partids : ",partids) print(" => partstack : ",partstack) # Run alignment command params = ali3d_base(projdata, get_im(paramsdict["refvol"]), \ ali3d_options, paramsdict["shrink"], mpi_comm = MPI_COMM_WORLD, log = log, \ nsoft = paramsdict["nsoft"], saturatecrit = paramsdict["saturatecrit"] ) del log, projdata # store params if(myid == main_node): line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"Executed successfully: ","ali3d_base_MPI %d"%paramsdict["nsoft"]," number of images:%7d"%len(params)) write_text_row(params, os.path.join(outputdir,"params-chunk%01d.txt"%procid) )
def mergeparfiles(i1,i2,io,p1,p2,po): # 1 - rescued # 2 - good old l1 = map(int, read_text_file( i1 )) l2 = map(int, read_text_file( i2 )) if(l1[0] == 0): write_text_file( l2, io) for ll in xrange(3): p = read_text_row(p2+"%01d.txt"%ll) write_text_row(2, po+"%01d.txt"%ll) else: t = l1 +l2 for i in xrange(len(t)): t[i] = [t[i],i] t.sort() write_text_file( [t[i][0] for i in xrange(len(t))], io) for ll in xrange(3): p = read_text_row(p1+"%01d.txt"%ll) + read_text_row(p2+"%01d.txt"%ll) write_text_row([p[t[i][1]] for i in xrange(len(t))], po+"%01d.txt"%ll) return
def output_volume(freqvol, resolut, apix, outvol, fsc, out_ang_res, nx, ny, nz, res_overall): outvol_ang = os.path.splitext(outvol)[0] + "_ang.hdf" outvol_shifted = os.path.splitext(outvol)[0] + "_shift.hdf" outvol_shifted_ang = os.path.splitext(outvol_shifted)[0] + "_ang.hdf" freqvol.write_image(outvol) if (out_ang_res): outAngResVol = makeAngRes(freqvol, nx, ny, nz, apix) outAngResVol.write_image(outvol_ang) if res_overall != -1.0: for ifreq in range(len(resolut)): if resolut[ifreq][0] > res_overall: break for jfreq in range(ifreq, len(resolut)): resolut[jfreq][1] = 0.0 data_freqvol = freqvol.get_3dview() mask = data_freqvol != 0 percentile_25 = numpy.percentile(data_freqvol[mask], 25) percentile_75 = numpy.percentile(data_freqvol[mask], 75) iqr = percentile_75 - percentile_25 mask_low_pass = data_freqvol > percentile_75 + 1.5 * iqr mask_high_pass = data_freqvol < percentile_25 - 1.5 * iqr mean_real = 1 / float( numpy.mean(data_freqvol[mask & mask_low_pass & mask_high_pass])) overall_res_real = 1 / float(res_overall) #mean_ang = options.apix / float(EMAN2_cppwrap.Util.infomask(freqvol, m, True)[0]) volume_out_real = makeAngRes(freqvol, nx, ny, nz, 1) volume_out_real += (overall_res_real - mean_real) volume_out = makeAngRes(volume_out_real, nx, ny, nz, 1, False) volume_out.write_image(outvol_shifted) if out_ang_res: outAngResVol = makeAngRes(freqvol, nx, ny, nz, apix) outAngResVol.write_image(outvol_shifted_ang) if (fsc != None): utilities.write_text_row(resolut, fsc)
def kernel(projections, stable_subset, target_threshold, options, minimal_subset_size, number_of_runs, number_of_winners, mpi_env, log, prefix=""): from multi_shc import multi_shc, find_common_subset_3 from utilities import wrap_mpi_bcast, write_text_row, write_text_file, wrap_mpi_gatherv, average_angles from itertools import combinations import os if log == None: from logger import Logger log = Logger() stable_subset = wrap_mpi_bcast(stable_subset, 0, mpi_env.main_comm) if mpi_env.main_rank == 0: log.add("Start ", number_of_runs, "* 3SHC") for i in xrange(number_of_runs): log.add("3SHC --> " + log.prefix + prefix + "_" + str(i)) completed_mshc = 0 params = [] while completed_mshc < number_of_runs: runs_to_do = min([(number_of_runs - completed_mshc), mpi_env.subcomms_count]) if mpi_env.subcomm_id < runs_to_do: out_dir = prefix + "_" + str(completed_mshc + mpi_env.subcomm_id) if mpi_env.sub_rank == 0: os.mkdir(log.prefix + out_dir) out_params, out_vol, out_peaks = multi_shc(projections, stable_subset, 3, options, mpi_env.sub_comm, log=log.sublog(out_dir + "/")) else: out_params = None if mpi_env.main_rank in mpi_env.subcomms_roots and mpi_env.subcomm_id < runs_to_do: params_temp = wrap_mpi_gatherv([out_params], 0, mpi_env.main_comm) else: params_temp = wrap_mpi_gatherv([], 0, mpi_env.main_comm) if mpi_env.main_rank == 0: params.extend(params_temp) completed_mshc += runs_to_do # find common subset if mpi_env.main_rank == 0: log.add("Calculate common subset") best_confs = [] largest_subset = [] largest_subset_error = 999.0 msg = "" for it in combinations(range(number_of_runs), number_of_winners): confs = list(it) input_params = [] for c in confs: input_params.append(params[c]) subset_thr, subset_size, err_thr, err_size = find_common_subset_3( input_params, target_threshold, minimal_subset_size, thresholds=True) msg += str(len(subset_size)) + "(" + str(err_size) + ") " if len(subset_size) > len(largest_subset) or ( len(subset_size) == len(largest_subset) and err_size < largest_subset_error): best_confs = confs largest_subset = subset_size largest_subset_error = err_size log.add(msg) subset = largest_subset threshold = largest_subset_error new_stable_subset = [] for i in subset: new_stable_subset.append(stable_subset[i]) log.add("Best solutions (winners): ", best_confs) average_params = [] for i in subset: temp = [params[j][i] for j in best_confs] average_params.append(average_angles(temp)) write_text_file(new_stable_subset, log.prefix + prefix + "_indexes.txt") write_text_row(average_params, log.prefix + prefix + "_params.txt") else: threshold = None new_stable_subset = None # broadcast threshold and new stable subset and exit threshold = wrap_mpi_bcast(threshold, 0, mpi_env.main_comm) new_stable_subset = wrap_mpi_bcast(new_stable_subset, 0, mpi_env.main_comm) return threshold, new_stable_subset
def main(): import os import sys from optparse import OptionParser from global_def import SPARXVERSION import global_def arglist = [] for arg in sys.argv: arglist.append( arg ) progname = os.path.basename(arglist[0]) usage2 = progname + """ inputfile outputfile [options] Functionalities: 1. Helicise input volume and save the result to output volume: sxhelicon_utils.py input_vol.hdf output_vol.hdf --helicise --dp=27.6 --dphi=166.5 --fract=0.65 --rmax=70 --rmin=1 --apix=1.84 --sym=D1 2. Helicise pdb file and save the result to a new pdb file: sxhelicon_utils.py input.pdb output.pdb --helicisepdb --dp=27.6 --dphi=166.5 --nrepeats --apix=1.84 3. Generate two lists of image indices used to split segment stack into halves for helical fsc calculation. sxhelicon_utils.py bdb:big_stack --hfsc='flst' --filament_attr=filament 4. Map of filament distribution in the stack sxhelicon_utils.py bdb:big_stack --filinfo=info.txt The output file will contain four columns: 1 2 3 4 first image number last image number number of images in the filament name 5. Predict segments' orientation parameters based on distances between segments and known helical symmetry sxhelicon_utils.py bdb:big_stack --predict_helical=helical_params.txt --dp=27.6 --dphi=166.5 --apix=1.84 6. Generate disks from filament based reconstructions: sxheader.py stk.hdf --params=xform.projection --import=params.txt # horatio active_refactoring Jy51i1EwmLD4tWZ9_00000_1 # sxheader.py stk.hdf --params=active --one mpirun -np 2 sxhelicon_utils.py stk.hdf --gendisk='bdb:disk' --ref_nx=100 --ref_ny=100 --ref_nz=200 --apix=1.84 --dp=27.6 --dphi=166.715 --fract=0.67 --rmin=0 --rmax=64 --function="[.,nofunc,helical3c]" --sym="c1" --MPI 7. Stack disks based on helical symmetry parameters sxhelicon_utils.py disk_to_stack.hdf --stackdisk=stacked_disks.hdf --dphi=166.5 --dp=27.6 --ref_nx=160 --ref_ny=160 --ref_nz=225 --apix=1.84 8. Helical symmetry search: mpirun -np 3 sxhelicon_utils.py volf0010.hdf outsymsearch --symsearch --dp=27.6 --dphi=166.715 --apix=1.84 --fract=0.65 --rmin=0 --rmax=92.0 --datasym=datasym.txt --dp_step=0.92 --ndp=3 --dphi_step=1.0 --ndphi=10 --MPI """ parser = OptionParser(usage2,version=SPARXVERSION) #parser.add_option("--ir", type="float", default= -1, help="inner radius for rotational correlation > 0 (set to 1) (Angstroms)") parser.add_option("--ou", type="float", default= -1, help="outer radius for rotational 2D correlation < int(nx/2)-1 (set to the radius of the particle) (Angstroms)") parser.add_option("--rs", type="int", default= 1, help="step between rings in rotational correlation >0 (set to 1)" ) parser.add_option("--xr", type="string", default= "4 2 1 1 1", help="range for translation search in x direction, search is +/-xr (Angstroms) ") parser.add_option("--txs", type="string", default= "1 1 1 0.5 0.25", help="step size of the translation search in x directions, search is -xr, -xr+ts, 0, xr-ts, xr (Angstroms)") parser.add_option("--delta", type="string", default= "10 6 4 3 2", help="angular step of reference projections") parser.add_option("--an", type="string", default= "-1", help="angular neighborhood for local searches") parser.add_option("--maxit", type="int", default= 30, help="maximum number of iterations performed for each angular step (set to 30) ") parser.add_option("--CTF", action="store_true", default=False, help="CTF correction") parser.add_option("--snr", type="float", default= 1.0, help="Signal-to-Noise Ratio of the data") parser.add_option("--MPI", action="store_true", default=False, help="use MPI version") #parser.add_option("--fourvar", action="store_true", default=False, help="compute Fourier variance") parser.add_option("--apix", type="float", default= -1.0, help="pixel size in Angstroms") parser.add_option("--dp", type="float", default= -1.0, help="delta z - translation in Angstroms") parser.add_option("--dphi", type="float", default= -1.0, help="delta phi - rotation in degrees") parser.add_option("--rmin", type="float", default= 0.0, help="minimal radius for hsearch (Angstroms)") parser.add_option("--rmax", type="float", default= 80.0, help="maximal radius for hsearch (Angstroms)") parser.add_option("--fract", type="float", default= 0.7, help="fraction of the volume used for helical search") parser.add_option("--sym", type="string", default= "c1", help="symmetry of the structure") parser.add_option("--function", type="string", default="helical", help="name of the reference preparation function") parser.add_option("--npad", type="int", default= 2, help="padding size for 3D reconstruction") parser.add_option("--debug", action="store_true", default=False, help="debug") parser.add_option("--volalixshift", action="store_true", default=False, help="Use volalixshift refinement") parser.add_option("--searchxshift", type="float", default= 0.0, help="search range for x-shift determination: +/- searchxshift (Angstroms)") parser.add_option("--nearby", type="float", default= 6.0, help="neighborhood within which to search for peaks in 1D ccf for x-shift search (Angstroms)") # filinfo parser.add_option( "--filinfo", type="string", default="", help="Store in an output text file infomration about distribution of filaments in the stack." ) # diskali parser.add_option("--diskali", action="store_true", default=False, help="volume alignment") parser.add_option("--zstep", type="float", default= 1, help="Step size for translational search along z (Angstroms)") # helicise parser.add_option("--helicise", action="store_true", default=False, help="helicise input volume and save results to output volume") parser.add_option("--hfsc", type="string", default="", help="Generate two lists of image indices used to split segment stack into halves for helical fsc calculation. The lists will be stored in two text files named using file_prefix with '_even' and '_odd' suffixes, respectively." ) parser.add_option("--filament_attr", type="string", default="filament", help="attribute under which filament identification is stored" ) parser.add_option("--predict_helical", type="string", default="", help="Generate projection parameters consistent with helical symmetry") # helicise pdb parser.add_option("--helicisepdb", action="store_true", default=False, help="Helicise pdb file and save the result to a new pdb file") parser.add_option("--nrepeats", type="int", default= 50, help="Number of time the helical symmetry will be applied to the input file") # input options for generating disks parser.add_option("--gendisk", type="string", default="", help="Name of file under which generated disks will be saved to") parser.add_option("--ref_nx", type="int", default= -1, help="nx=ny volume size" ) parser.add_option("--ref_nz", type="int", default= -1, help="nz volume size - computed disks will be nx x ny x rise/apix" ) parser.add_option("--new_pixel_size", type="float", default= -1, help="desired pixel size of the output disks. The default is -1, in which case there is no resampling (unless --match_pixel_rise flag is True).") parser.add_option("--maxerror", type="float", default= 0.1, help="proportional to the maximum amount of error to tolerate between (dp/new_pixel_size) and int(dp/new_pixel_size ), where new_pixel_size is the pixel size calculated when the option --match_pixel_rise flag is True.") parser.add_option("--match_pixel_rise", action="store_true", default=False, help="calculate new pixel size such that the rise is approximately integer number of pixels given the new pixel size. This will be the pixel size of the output disks.") # get consistency parser.add_option("--consistency", type="string", default="", help="Name of parameters to get consistency statistics for") parser.add_option("--phithr", type="float", default= 2.0, help="phi threshold for consistency check") parser.add_option("--ythr", type="float", default= 2.0, help="y threshold (in Angstroms) for consistency check") parser.add_option("--segthr", type="int", default= 3, help="minimum number of segments/filament for consistency check") # stack disks parser.add_option("--stackdisk", type="string", default="", help="Name of file under which output volume will be saved to.") parser.add_option("--ref_ny", type="int", default=-1, help="ny of output volume size. Default is ref_nx" ) # symmetry search parser.add_option("--symsearch", action="store_true", default=False, help="Do helical symmetry search." ) parser.add_option("--ndp", type="int", default= 12, help="In symmetrization search, number of delta z steps equals to 2*ndp+1") parser.add_option("--ndphi", type="int", default= 12, help="In symmetrization search, number of dphi steps equals to 2*ndphi+1") parser.add_option("--dp_step", type="float", default= 0.1, help="delta z step for symmetrization [Angstroms] (default 0.1)") parser.add_option("--dphi_step", type="float", default= 0.1, help="dphi step for symmetrization [degrees] (default 0.1)") parser.add_option("--datasym", type="string", default="datasym.txt", help="symdoc") parser.add_option("--symdoc", type="string", default="", help="text file containing helical symmetry parameters dp and dphi") # filament statistics in the stack (options, args) = parser.parse_args(arglist[1:]) if len(args) < 1 or len(args) > 5: print "Various helical reconstruction related functionalities: " + usage2 print "Please run '" + progname + " -h' for detailed options" else: if len(options.hfsc) > 0: if len(args) != 1: print "Incorrect number of parameters" sys.exit() from applications import imgstat_hfsc imgstat_hfsc( args[0], options.hfsc, options.filament_attr) sys.exit() elif len(options.filinfo) > 0: if len(args) != 1: print "Incorrect number of parameters" sys.exit() from EMAN2 import EMUtil filams = EMUtil.get_all_attributes(args[0], "filament") ibeg = 0 filcur = filams[0] n = len(filams) inf = [] i = 1 while( i <= n): if(i < n): fis = filams[i] else: fis = "" if( fis != filcur ): iend = i-1 inf.append([ibeg,iend,iend-ibeg+1,filcur]) ibeg = i filcur = fis i += 1 from utilities import write_text_row write_text_row(inf, options.filinfo) sys.exit() if len(options.stackdisk) > 0: if len(args) != 1: print "Incorrect number of parameters" sys.exit() dpp = (float(options.dp)/options.apix) rise = int(dpp) if(abs(float(rise) - dpp)>1.0e-3): print " dpp has to be integer multiplicity of the pixel size" sys.exit() from utilities import get_im v = get_im(args[0]) from applications import stack_disks ref_ny = options.ref_ny if ref_ny < 0: ref_ny = options.ref_nx sv = stack_disks(v, options.ref_nx, ref_ny, options.ref_nz, options.dphi, rise) sv.write_image(options.stackdisk) sys.exit() if len(options.consistency) > 0: if len(args) != 1: print "Incorrect number of parameters" sys.exit() from development import consistency_params consistency_params(args[0], options.consistency, options.dphi, options.dp, options.apix,phithr=options.phithr, ythr=options.ythr, THR=options.segthr) sys.exit() rminp = int((float(options.rmin)/options.apix) + 0.5) rmaxp = int((float(options.rmax)/options.apix) + 0.5) from utilities import get_input_from_string, get_im xr = get_input_from_string(options.xr) txs = get_input_from_string(options.txs) irp = 1 if options.ou < 0: oup = -1 else: oup = int( (options.ou/options.apix) + 0.5) xrp = '' txsp = '' for i in xrange(len(xr)): xrp += " "+str(float(xr[i])/options.apix) for i in xrange(len(txs)): txsp += " "+str(float(txs[i])/options.apix) searchxshiftp = int( (options.searchxshift/options.apix) + 0.5) nearbyp = int( (options.nearby/options.apix) + 0.5) zstepp = int( (options.zstep/options.apix) + 0.5) if options.MPI: from mpi import mpi_init, mpi_finalize sys.argv = mpi_init(len(sys.argv), sys.argv) if len(options.predict_helical) > 0: if len(args) != 1: print "Incorrect number of parameters" sys.exit() if options.dp < 0: print "Helical symmetry paramter rise --dp should not be negative" sys.exit() from applications import predict_helical_params predict_helical_params(args[0], options.dp, options.dphi, options.apix, options.predict_helical) sys.exit() if options.helicise: if len(args) != 2: print "Incorrect number of parameters" sys.exit() if options.dp < 0: print "Helical symmetry paramter rise --dp should not be negative" sys.exit() from utilities import get_im, sym_vol vol = get_im(args[0]) vol = sym_vol(vol, options.sym) hvol = vol.helicise(options.apix, options.dp, options.dphi, options.fract, rmaxp, rminp) hvol = sym_vol(hvol, options.sym) hvol.write_image(args[1]) sys.exit() if options.helicisepdb: if len(args) != 2: print "Incorrect number of parameters" sys.exit() if options.dp < 0: print "Helical symmetry paramter rise --dp should not be negative" sys.exit() from math import cos, sin, radians from copy import deepcopy import numpy from numpy import zeros,dot,float32 dp = options.dp dphi = options.dphi nperiod = options.nrepeats infile =open(args[0],"r") pall = infile.readlines() infile.close() p = [] pos = [] lkl = -1 for i in xrange( len(pall) ): if( (pall[i])[:4] == 'ATOM'): if( lkl == -1 ): lkl = i p.append( pall[i] ) pos.append(i) n = len(p) X = zeros( (3,len(p) ), dtype=float32 ) X_new = zeros( (3,len(p) ), dtype=float32 ) for i in xrange( len(p) ): element = deepcopy( p[i] ) X[0,i]=float(element[30:38]) X[1,i]=float(element[38:46]) X[2,i]=float(element[46:54]) pnew = [] for j in xrange(-nperiod, nperiod+1): for i in xrange( n ): pnew.append( deepcopy(p[i]) ) dphi = radians(dphi) m = zeros( (3,3 ), dtype=float32 ) t = zeros( (3,1 ), dtype=float32 ) m[2][2] = 1.0 t[0,0] = 0.0 t[1,0] = 0.0 for j in xrange(-nperiod, nperiod+1): if j != 0: rd = j*dphi m[0][0] = cos(rd) m[0][1] = sin(rd) m[1][0] = -m[0][1] m[1][1] = m[0][0] t[2,0] = j*dp X_new = dot(m, X) + t for i in xrange( n ): pnew[j*n+i] = pnew[j*n+i][:30] + "%8.3f"%( float(X_new[0,i]) )+"%8.3f"%( float(X_new[1,i]) )+"%8.3f"%( float(X_new[2,i]) ) + pnew[j*n+i][54:] outfile=open(args[1],"w") outfile.writelines(pall[0:lkl]) outfile.writelines(pnew) outfile.writelines("END\n") outfile.close() sys.exit() if options.volalixshift: if options.maxit > 1: print "Inner iteration for x-shift determinatin is restricted to 1" sys.exit() if len(args) < 4: mask = None else: mask = args[3] from applications import volalixshift_MPI global_def.BATCH = True volalixshift_MPI(args[0], args[1], args[2], searchxshiftp, options.apix, options.dp, options.dphi, options.fract, rmaxp, rminp, mask, options.maxit, options.CTF, options.snr, options.sym, options.function, options.npad, options.debug, nearbyp) global_def.BATCH = False if options.diskali: #if options.maxit > 1: # print "Inner iteration for disk alignment is restricted to 1" # sys.exit() if len(args) < 4: mask = None else: mask = args[3] global_def.BATCH = True if(options.sym[:1] == "d" or options.sym[:1] == "D" ): from development import diskaliD_MPI diskaliD_MPI(args[0], args[1], args[2], mask, options.dp, options.dphi, options.apix, options.function, zstepp, options.fract, rmaxp, rminp, options.CTF, options.maxit, options.sym) else: from applications import diskali_MPI diskali_MPI(args[0], args[1], args[2], mask, options.dp, options.dphi, options.apix, options.function, zstepp, options.fract, rmaxp, rminp, options.CTF, options.maxit, options.sym) global_def.BATCH = False if options.symsearch: if len(options.symdoc) < 1: if options.dp < 0 or options.dphi < 0: print "Enter helical symmetry parameters either using --symdoc or --dp and --dphi" sys.exit() if options.dp < 0 or options.dphi < 0: # read helical symmetry parameters from symdoc from utilities import read_text_row hparams=read_text_row(options.symdoc) dp = hparams[0][0] dphi = hparams[0][1] else: dp = options.dp dphi = options.dphi from applications import symsearch_MPI if len(args) < 3: mask = None else: mask= args[2] global_def.BATCH = True symsearch_MPI(args[0], args[1], mask, dp, options.ndp, options.dp_step, dphi, options.ndphi, options.dphi_step, rminp, rmaxp, options.fract, options.sym, options.function, options.datasym, options.apix, options.debug) global_def.BATCH = False elif len(options.gendisk)> 0: from applications import gendisks_MPI global_def.BATCH = True if len(args) == 1: mask3d = None else: mask3d = args[1] if options.dp < 0: print "Helical symmetry paramter rise --dp must be explictly set!" sys.exit() gendisks_MPI(args[0], mask3d, options.ref_nx, options.apix, options.dp, options.dphi, options.fract, rmaxp, rminp, options.CTF, options.function, options.sym, options.gendisk, options.maxerror, options.new_pixel_size, options.match_pixel_rise) global_def.BATCH = False if options.MPI: from mpi import mpi_finalize mpi_finalize()
def compute_fscs(stack, outputdir, chunkname, newgoodname, fscoutputdir, doit, keepchecking, nproc, myid, main_node): # Compute reconstructions per group from good particles only to get FSC curves # We will compute two FSC curves - from not averaged parameters and from averaged parameters # So, we have to build two sets: # not averaged (A2+C3) versus (B0+D5) # averaged (A0+C1) versus (B3+D4) # This requires pulling good subsets given by goodX*; I am not sure why good, sxconsistency above produced newgood text files. # Otherwise, I am not sure what newbad will contain. # Input that should vary: # "bdb:"+os.path.join(outputdir,"chunk%01d%01d"%(procid,i)) # os.path.join(outputdir,"newgood%01d.txt"%procid) # Output should be in a separate directory "fscoutputdir" if(myid == main_node): if keepchecking: if(os.path.exists(fscoutputdir)): doit = 0 print("Directory ",fscoutputdir," exists!") else: doit = 1 keepchecking = False else: doit = 1 if doit: cmd = "{} {}".format("mkdir", fscoutputdir) cmdexecute(cmd) mpi_barrier(MPI_COMM_WORLD) # not averaged doit, keepchecking = checkstep(os.path.join(fscoutputdir,"volfscn0.hdf"), keepchecking, myid, main_node) if doit: if(myid == main_node): # A2+C3 # indices write_text_file( \ map(int, read_text_file(os.path.join(outputdir,chunkname+"0.txt")))+map(int, read_text_file(os.path.join(outputdir,chunkname+"2.txt"))), \ os.path.join(fscoutputdir,"chunkfn0.txt")) # params write_text_row( \ read_text_row(os.path.join(outputdir,newgoodname+"02.txt"))+read_text_row(os.path.join(outputdir,newgoodname+"22.txt")), \ os.path.join(fscoutputdir,"params-chunkfn0.txt")) mpi_barrier(MPI_COMM_WORLD) projdata = getindexdata(stack, os.path.join(fscoutputdir,"chunkfn0.txt"), os.path.join(fscoutputdir,"params-chunkfn0.txt"), myid, nproc) if ali3d_options.CTF: vol = recons3d_4nn_ctf_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) else: vol = recons3d_4nn_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) del projdata if(myid == main_node): vol.write_image(os.path.join(fscoutputdir,"volfscn0.hdf")) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"Executed successfully: 3D reconstruction of", os.path.join(fscoutputdir,"volfscn0.hdf")) del vol doit, keepchecking = checkstep(os.path.join(fscoutputdir,"volfscn1.hdf"), keepchecking, myid, main_node) if doit: if(myid == main_node): # B0+D5 # indices write_text_file( \ map(int, read_text_file(os.path.join(outputdir,chunkname+"1.txt")))+map(int, read_text_file(os.path.join(outputdir,chunkname+"3.txt"))), \ os.path.join(fscoutputdir,"chunkfn1.txt")) # params write_text_row( \ read_text_row(os.path.join(outputdir,newgoodname+"10.txt"))+read_text_row(os.path.join(outputdir,newgoodname+"32.txt")), \ os.path.join(fscoutputdir,"params-chunkfn1.txt")) mpi_barrier(MPI_COMM_WORLD) projdata = getindexdata(stack, os.path.join(fscoutputdir,"chunkfn1.txt"), os.path.join(fscoutputdir,"params-chunkfn1.txt"), myid, nproc) if ali3d_options.CTF: vol = recons3d_4nn_ctf_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) else: vol = recons3d_4nn_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) del projdata if(myid == main_node): vol.write_image(os.path.join(fscoutputdir,"volfscn1.hdf")) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"Executed successfully: 3D reconstruction of", os.path.join(fscoutputdir,"volfscn1.hdf")) del vol # averaged doit, keepchecking = checkstep(os.path.join(fscoutputdir,"volfsca0.hdf"), keepchecking, myid, main_node) if doit: if(myid == main_node): # A0+C1 # indices write_text_file( \ map(int, read_text_file(os.path.join(outputdir,chunkname+"0.txt")))+map(int, read_text_file(os.path.join(outputdir,chunkname+"2.txt"))), \ os.path.join(fscoutputdir,"chunkfa0.txt")) # params write_text_row( \ read_text_row(os.path.join(outputdir,newgoodname+"00.txt"))+read_text_row(os.path.join(outputdir,newgoodname+"20.txt")), \ os.path.join(fscoutputdir,"params-chunkfa0.txt")) mpi_barrier(MPI_COMM_WORLD) projdata = getindexdata(stack, os.path.join(fscoutputdir,"chunkfa0.txt"), os.path.join(fscoutputdir,"params-chunkfa0.txt"), myid, nproc) if ali3d_options.CTF: vol = recons3d_4nn_ctf_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) else: vol = recons3d_4nn_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) del projdata if(myid == main_node): vol.write_image(os.path.join(fscoutputdir,"volfsca0.hdf")) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"Executed successfully: 3D reconstruction of", os.path.join(fscoutputdir,"volfsca0.hdf")) del vol doit, keepchecking = checkstep(os.path.join(fscoutputdir,"volfsca1.hdf"), keepchecking, myid, main_node) if doit: if(myid == main_node): # B3+D4 write_text_file( \ map(int, read_text_file(os.path.join(outputdir,chunkname+"1.txt")))+map(int, read_text_file(os.path.join(outputdir,chunkname+"3.txt"))), \ os.path.join(fscoutputdir,"chunkfa1.txt")) # params write_text_row( \ read_text_row(os.path.join(outputdir,newgoodname+"11.txt"))+read_text_row(os.path.join(outputdir,newgoodname+"31.txt")), \ os.path.join(fscoutputdir,"params-chunkfa1.txt")) mpi_barrier(MPI_COMM_WORLD) projdata = getindexdata(stack, os.path.join(fscoutputdir,"chunkfa1.txt"), os.path.join(fscoutputdir,"params-chunkfa1.txt"), myid, nproc) if ali3d_options.CTF: vol = recons3d_4nn_ctf_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) else: vol = recons3d_4nn_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) del projdata if(myid == main_node): vol.write_image(os.path.join(fscoutputdir,"volfsca1.hdf")) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"Executed successfully: 3D reconstruction of", os.path.join(fscoutputdir,"volfsca1.hdf")) del vol # Get updated FSC curves if(myid == main_node): if(ali3d_options.mask3D is None): mask = model_circle(radi,nnxo,nnxo,nnxo) else: mask = get_im(ali3d_options.mask3D) if keepchecking: if(os.path.exists(os.path.join(fscoutputdir,"fscn.txt"))): doit = 0 else: doit = 1 keepchecking = False else: doit = 1 if doit: fsc(get_im(os.path.join(fscoutputdir,"volfscn0.hdf"))*mask,\ get_im(os.path.join(fscoutputdir,"volfscn1.hdf"))*mask,\ 1.0,os.path.join(fscoutputdir,"fscn.txt") ) if keepchecking: if(os.path.exists(os.path.join(fscoutputdir,"fsca.txt"))): doit = 0 else: doit = 1 keepchecking = False else: doit = 1 if doit: fsc(get_im(os.path.join(fscoutputdir,"volfsca0.hdf"))*mask,\ get_im(os.path.join(fscoutputdir,"volfsca1.hdf"))*mask,\ 1.0,os.path.join(fscoutputdir,"fsca.txt") ) nfsc = read_text_file(os.path.join(fscoutputdir,"fscn.txt") ,-1) currentres = 0.5 ns = len(nfsc[1]) for i in xrange(1,ns-1): if ( (2*nfsc[1][i]/(1.0+nfsc[1][i]) ) < 0.5): currentres = nfsc[0][i-1] break print(" Current resolution ",i,currentres) else: currentres = 0.0 currentres = bcast_number_to_all(currentres, source_node = main_node) if(currentres < 0.0): if(myid == main_node): print(" Something wrong with the resolution, cannot continue") mpi_finalize() exit() mpi_barrier(MPI_COMM_WORLD) return currentres, doit, keepchecking
def main(): from utilities import get_input_from_string progname = os.path.basename(sys.argv[0]) usage = progname + " stack output_average --radius=particle_radius --xr=xr --yr=yr --ts=ts --thld_err=thld_err --num_ali=num_ali --fl=fl --aa=aa --CTF --verbose --stables" parser = OptionParser(usage,version=SPARXVERSION) parser.add_option("--radius", type="int", default=-1, help=" particle radius for alignment") parser.add_option("--xr", type="string" , default="2 1", help="range for translation search in x direction, search is +/xr (default 2,1)") parser.add_option("--yr", type="string" , default="-1", help="range for translation search in y direction, search is +/yr (default = same as xr)") parser.add_option("--ts", type="string" , default="1 0.5", help="step size of the translation search in both directions, search is -xr, -xr+ts, 0, xr-ts, xr, can be fractional (default: 1,0.5)") parser.add_option("--thld_err", type="float", default=0.75, help="threshld of pixel error (default = 0.75)") parser.add_option("--num_ali", type="int", default=5, help="number of alignments performed for stability (default = 5)") parser.add_option("--maxit", type="int", default=30, help="number of iterations for each xr (default = 30)") parser.add_option("--fl", type="float" , default=0.3, help="cut-off frequency of hyperbolic tangent low-pass Fourier filter (default = 0.3)") parser.add_option("--aa", type="float" , default=0.2, help="fall-off of hyperbolic tangent low-pass Fourier filter (default = 0.2)") parser.add_option("--CTF", action="store_true", default=False, help="Use CTF correction during the alignment ") parser.add_option("--verbose", action="store_true", default=False, help="print individual pixel error (default = False)") parser.add_option("--stables", action="store_true", default=False, help="output the stable particles number in file (default = False)") parser.add_option("--method", type="string" , default=" ", help="SHC (standard method is default when flag is ommitted)") (options, args) = parser.parse_args() if len(args) != 1 and len(args) != 2: print "usage: " + usage print "Please run '" + progname + " -h' for detailed options" else: if global_def.CACHE_DISABLE: from utilities import disable_bdb_cache disable_bdb_cache() from applications import within_group_refinement, ali2d_ras from pixel_error import multi_align_stability from utilities import write_text_file, write_text_row global_def.BATCH = True xrng = get_input_from_string(options.xr) if options.yr == "-1": yrng = xrng else : yrng = get_input_from_string(options.yr) step = get_input_from_string(options.ts) class_data = EMData.read_images(args[0]) nx = class_data[0].get_xsize() ou = options.radius num_ali = options.num_ali if ou == -1: ou = nx/2-2 from utilities import model_circle, get_params2D, set_params2D mask = model_circle(ou, nx, nx) if options.CTF : from filter import filt_ctf for im in xrange(len(class_data)): # Flip phases class_data[im] = filt_ctf(class_data[im], class_data[im].get_attr("ctf"), binary=1) for im in class_data: im.set_attr("previousmax", -1.0e10) try: t = im.get_attr("xform.align2d") # if they are there, no need to set them! except: try: t = im.get_attr("xform.projection") d = t.get_params("spider") set_params2D(im, [0.0, -d["tx"], -d["ty"], 0, 1.0]) except: set_params2D(im, [0.0, 0.0, 0.0, 0, 1.0]) all_ali_params = [] for ii in xrange(num_ali): ali_params = [] if options.verbose: ALPHA = [] SX = [] SY = [] MIRROR = [] if( xrng[0] == 0.0 and yrng[0] == 0.0 ): avet = ali2d_ras(class_data, randomize = True, ir = 1, ou = ou, rs = 1, step = 1.0, dst = 90.0, \ maxit = options.maxit, check_mirror = True, FH=options.fl, FF=options.aa) else: avet = within_group_refinement(class_data, mask, True, 1, ou, 1, xrng, yrng, step, 90.0, \ maxit = options.maxit, FH=options.fl, FF=options.aa, method = options.method) from utilities import info #print " avet ",info(avet) for im in class_data: alpha, sx, sy, mirror, scale = get_params2D(im) ali_params.extend([alpha, sx, sy, mirror]) if options.verbose: ALPHA.append(alpha) SX.append(sx) SY.append(sy) MIRROR.append(mirror) all_ali_params.append(ali_params) if options.verbose: write_text_file([ALPHA, SX, SY, MIRROR], "ali_params_run_%d"%ii) """ avet = class_data[0] from utilities import read_text_file all_ali_params = [] for ii in xrange(5): temp = read_text_file( "ali_params_run_%d"%ii,-1) uuu = [] for k in xrange(len(temp[0])): uuu.extend([temp[0][k],temp[1][k],temp[2][k],temp[3][k]]) all_ali_params.append(uuu) """ stable_set, mir_stab_rate, pix_err = multi_align_stability(all_ali_params, 0.0, 10000.0, options.thld_err, options.verbose, 2*ou+1) print "%4s %20s %20s %20s %30s %6.2f"%("", "Size of set", "Size of stable set", "Mirror stab rate", "Pixel error prior to pruning the set above threshold of",options.thld_err) print "Average stat: %10d %20d %20.2f %15.2f"%( len(class_data), len(stable_set), mir_stab_rate, pix_err) if( len(stable_set) > 0): if options.stables: stab_mem = [[0,0.0,0] for j in xrange(len(stable_set))] for j in xrange(len(stable_set)): stab_mem[j] = [int(stable_set[j][1]), stable_set[j][0], j] write_text_row(stab_mem, "stable_particles.txt") stable_set_id = [] particle_pixerr = [] for s in stable_set: stable_set_id.append(s[1]) particle_pixerr.append(s[0]) from fundamentals import rot_shift2D avet.to_zero() l = -1 print "average parameters: angle, x-shift, y-shift, mirror" for j in stable_set_id: l += 1 print " %4d %4d %12.2f %12.2f %12.2f %1d"%(l,j, stable_set[l][2][0], stable_set[l][2][1], stable_set[l][2][2], int(stable_set[l][2][3])) avet += rot_shift2D(class_data[j], stable_set[l][2][0], stable_set[l][2][1], stable_set[l][2][2], stable_set[l][2][3] ) avet /= (l+1) avet.set_attr('members', stable_set_id) avet.set_attr('pix_err', pix_err) avet.set_attr('pixerr', particle_pixerr) avet.write_image(args[1]) global_def.BATCH = False
def doXfiles(path, source = "chunk", inparams = "params", params = "params", dest = "X"): # will produce X*.txt and paramsX*.txt # Generate six Xfiles from four chunks and export parameters. This is hardwired as it is always done in the same way # AB # indices write_text_file( \ map(int, read_text_file(os.path.join(path,source+"0.txt")))+map(int, read_text_file(os.path.join(path,source+"1.txt"))), \ os.path.join(path,dest+"0.txt")) # params write_text_row( \ read_text_row(os.path.join(path,inparams+"00.txt"))+read_text_row(os.path.join(path,inparams+"10.txt")), \ os.path.join(path,params+dest+"0.txt")) # AC write_text_file( \ map(int, read_text_file(os.path.join(path,source+"0.txt")))+map(int, read_text_file(os.path.join(path,source+"2.txt"))), \ os.path.join(path,dest+"1.txt")) write_text_row( \ read_text_row(os.path.join(path,inparams+"01.txt"))+read_text_row(os.path.join(path,inparams+"20.txt")), \ os.path.join(path,params+dest+"1.txt")) # AD write_text_file( \ map(int, read_text_file(os.path.join(path,source+"0.txt")))+map(int, read_text_file(os.path.join(path,source+"3.txt"))), \ os.path.join(path,dest+"2.txt")) write_text_row( \ read_text_row(os.path.join(path,inparams+"02.txt"))+read_text_row(os.path.join(path,inparams+"30.txt")), \ os.path.join(path,params+dest+"2.txt")) # BC write_text_file( \ map(int, read_text_file(os.path.join(path,source+"1.txt")))+map(int, read_text_file(os.path.join(path,source+"2.txt"))), \ os.path.join(path,dest+"3.txt")) write_text_row( \ read_text_row(os.path.join(path,inparams+"11.txt"))+read_text_row(os.path.join(path,inparams+"21.txt")), \ os.path.join(path,params+dest+"3.txt")) # BD write_text_file( \ map(int, read_text_file(os.path.join(path,source+"1.txt")))+map(int, read_text_file(os.path.join(path,source+"3.txt"))), \ os.path.join(path,dest+"4.txt")) write_text_row( \ read_text_row(os.path.join(path,inparams+"12.txt"))+read_text_row(os.path.join(path,inparams+"31.txt")), \ os.path.join(path,params+dest+"4.txt")) # CD write_text_file( \ map(int, read_text_file(os.path.join(path,source+"2.txt")))+map(int, read_text_file(os.path.join(path,source+"3.txt"))), \ os.path.join(path,dest+"5.txt")) write_text_row( \ read_text_row(os.path.join(path,inparams+"22.txt"))+read_text_row(os.path.join(path,inparams+"32.txt")), \ os.path.join(path,params+dest+"5.txt")) return
def main(): import sys import os import math import random import pyemtbx.options import time from random import random, seed, randint from optparse import OptionParser progname = os.path.basename(sys.argv[0]) usage = progname + """ [options] <inputfile> <outputfile> Forms chains of 2D images based on their similarities. Functionality: 4. Order a 2-D stack of image based on pair-wise similarity (computed as a cross-correlation coefficent). Options 1-3 require image stack to be aligned. The program will apply orientation parameters if present in headers. The ways to use the program: 4.1 Use option initial to specify which image will be used as an initial seed to form the chain. sxprocess.py input_stack.hdf output_stack.hdf --initial=23 --radius=25 4.2 If options initial is omitted, the program will determine which image best serves as initial seed to form the chain sxprocess.py input_stack.hdf output_stack.hdf --radius=25 4.3 Use option circular to form a circular chain. sxprocess.py input_stack.hdf output_stack.hdf --circular--radius=25 4.4 New circular code based on pairwise alignments sxprocess.py aclf.hdf chain.hdf circle.hdf --align --radius=25 --xr=2 --pairwiseccc=lcc.txt 4.5 Circular ordering based on pairwise alignments sxprocess.py vols.hdf chain.hdf mask.hdf --dd --radius=25 """ parser = OptionParser(usage,version=SPARXVERSION) parser.add_option("--dd", action="store_true", help="Circular ordering without adjustment of orientations", default=False) parser.add_option("--circular", action="store_true", help="Select circular ordering (first image has to be similar to the last)", default=False) parser.add_option("--align", action="store_true", help="Compute all pairwise alignments and for the table of their similarities find the best chain", default=False) parser.add_option("--initial", type="int", default=-1, help="Specifies which image will be used as an initial seed to form the chain. (default = 0, means the first image)") parser.add_option("--radius", type="int", default=-1, help="Radius of a circular mask for similarity based ordering") # import params for 2D alignment parser.add_option("--ou", type="int", default=-1, help="outer radius for 2D alignment < nx/2-1 (set to the radius of the particle)") parser.add_option("--xr", type="int", default=0, help="range for translation search in x direction, search is +/xr (0)") parser.add_option("--yr", type="int", default=0, help="range for translation search in y direction, search is +/yr (0)") #parser.add_option("--nomirror", action="store_true", default=False, help="Disable checking mirror orientations of images (default False)") parser.add_option("--pairwiseccc", type="string", default= None, help="Input/output pairwise ccc file") (options, args) = parser.parse_args() global_def.BATCH = True if options.dd: nargs = len(args) if nargs != 3: print "must provide name of input and two output files!" return stack = args[0] new_stack = args[1] from utilities import model_circle from statistics import ccc from statistics import mono lend = EMUtil.get_image_count(stack) lccc = [None]*(lend*(lend-1)/2) for i in xrange(lend-1): v1 = get_im( stack, i ) if( i == 0 and nargs == 2): nx = v1.get_xsize() ny = v1.get_ysize() nz = v1.get_ysize() if options.ou < 1 : radius = nx//2-2 else: radius = options.ou mask = model_circle(radius, nx, ny, nz) else: mask = get_im(args[2]) for j in xrange(i+1, lend): lccc[mono(i,j)] = [ccc(v1, get_im( stack, j ), mask), 0] order = tsp(lccc) if(len(order) != lend): print " problem with data length" from sys import exit exit() print "Total sum of cccs :",TotalDistance(order, lccc) print "ordering :",order for i in xrange(lend): get_im(stack, order[i]).write_image( new_stack, i ) elif options.align: nargs = len(args) if nargs != 3: print "must provide name of input and two output files!" return from utilities import get_params2D, model_circle from fundamentals import rot_shift2D from statistics import ccc from time import time from alignment import align2d, align2d_scf from multi_shc import mult_transform stack = args[0] new_stack = args[1] d = EMData.read_images(stack) """ # will align anyway try: ttt = d[0].get_attr('xform.params2d') for i in xrange(len(d)): alpha, sx, sy, mirror, scale = get_params2D(d[i]) d[i] = rot_shift2D(d[i], alpha, sx, sy, mirror) except: pass """ nx = d[0].get_xsize() ny = d[0].get_ysize() if options.ou < 1 : radius = nx//2-2 else: radius = options.ou mask = model_circle(radius, nx, ny) if(options.xr < 0): xrng = 0 else: xrng = options.xr if(options.yr < 0): yrng = xrng else: yrng = options.yr initial = max(options.initial, 0) from statistics import mono lend = len(d) lccc = [None]*(lend*(lend-1)/2) from utilities import read_text_row if options.pairwiseccc == None or not os.path.exists(options.pairwiseccc) : st = time() for i in xrange(lend-1): for j in xrange(i+1, lend): # j>i meaning mono entry (i,j) or (j,i) indicates T i->j (from smaller index to larger) #alpha, sx, sy, mir, peak = align2d(d[i],d[j], xrng, yrng, step=options.ts, first_ring=options.ir, last_ring=radius, mode = "F") alpha, sx, sy, mir, peak = align2d_scf(d[i],d[j], xrng, yrng, ou=radius) lccc[mono(i,j)] = [ccc(d[j], rot_shift2D(d[i], alpha, sx, sy, mir, 1.0), mask), alpha, sx, sy, mir] #print " %4d %10.1f"%(i,time()-st) if(not os.path.exists(options.pairwiseccc)): from utilities import write_text_row write_text_row([[initial,0,0,0,0]]+lccc,options.pairwiseccc) elif(os.path.exists(options.pairwiseccc)): lccc = read_text_row(options.pairwiseccc) initial = int(lccc[0][0] + 0.1) del lccc[0] for i in xrange(len(lccc)): T = Transform({"type":"2D","alpha":lccc[i][1],"tx":lccc[i][2],"ty":lccc[i][3],"mirror":int(lccc[i][4]+0.1)}) lccc[i] = [lccc[i][0],T] tdummy = Transform({"type":"2D"}) maxsum = -1.023 for m in xrange(0,lend):#initial, initial+1): indc = range( lend ) lsnake = [[m, tdummy, 0.0]] del indc[m] lsum = 0.0 while len(indc) > 1: maxcit = -111. for i in xrange(len(indc)): cuc = lccc[mono(indc[i], lsnake[-1][0])][0] if cuc > maxcit: maxcit = cuc qi = indc[i] # Here we need transformation from the current to the previous, # meaning indc[i] -> lsnake[-1][0] T = lccc[mono(indc[i], lsnake[-1][0])][1] # If direction is from larger to smaller index, the transformation has to be inverted if( indc[i] > lsnake[-1][0] ): T = T.inverse() lsnake.append([qi,T, maxcit]) lsum += maxcit del indc[indc.index(qi)] T = lccc[mono(indc[-1], lsnake[-1][0])][1] if( indc[-1] > lsnake[-1][0]): T = T.inverse() lsnake.append([indc[-1], T, lccc[mono(indc[-1], lsnake[-1][0])][0]]) print " initial image and lsum ",m,lsum #print lsnake if(lsum > maxsum): maxsum = lsum init = m snake = [lsnake[i] for i in xrange(lend)] print " Initial image selected : ",init,maxsum," ",TotalDistance([snake[m][0] for m in xrange(lend)], lccc) #for q in snake: print q from copy import deepcopy trans=deepcopy([snake[i][1] for i in xrange(len(snake))]) print [snake[i][0] for i in xrange(len(snake))]
def main(): import sys import os import math import random import pyemtbx.options import time from random import random, seed, randint from optparse import OptionParser progname = os.path.basename(sys.argv[0]) usage = progname + """ [options] <inputfile> <outputfile> Forms chains of 2D images based on their similarities. Functionality: 4. Order a 2-D stack of image based on pair-wise similarity (computed as a cross-correlation coefficent). Options 1-3 require image stack to be aligned. The program will apply orientation parameters if present in headers. The ways to use the program: 4.1 Use option initial to specify which image will be used as an initial seed to form the chain. sxprocess.py input_stack.hdf output_stack.hdf --initial=23 --radius=25 4.2 If options initial is omitted, the program will determine which image best serves as initial seed to form the chain sxprocess.py input_stack.hdf output_stack.hdf --radius=25 4.3 Use option circular to form a circular chain. sxprocess.py input_stack.hdf output_stack.hdf --circular--radius=25 4.4 New circular code based on pairwise alignments sxprocess.py aclf.hdf chain.hdf circle.hdf --align --radius=25 --xr=2 --pairwiseccc=lcc.txt 4.5 Circular ordering based on pairwise alignments sxprocess.py vols.hdf chain.hdf mask.hdf --dd --radius=25 """ parser = OptionParser(usage,version=SPARXVERSION) parser.add_option("--dd", action="store_true", help="Circular ordering without adjustment of orientations", default=False) parser.add_option("--circular", action="store_true", help="Select circular ordering (first image has to be similar to the last)", default=False) parser.add_option("--align", action="store_true", help="Compute all pairwise alignments and for the table of their similarities find the best chain", default=False) parser.add_option("--initial", type="int", default=-1, help="Specifies which image will be used as an initial seed to form the chain. (default = 0, means the first image)") parser.add_option("--radius", type="int", default=-1, help="Radius of a circular mask for similarity based ordering") # import params for 2D alignment parser.add_option("--ou", type="int", default=-1, help="outer radius for 2D alignment < nx/2-1 (set to the radius of the particle)") parser.add_option("--xr", type="int", default=0, help="range for translation search in x direction, search is +/xr (0)") parser.add_option("--yr", type="int", default=0, help="range for translation search in y direction, search is +/yr (0)") #parser.add_option("--nomirror", action="store_true", default=False, help="Disable checking mirror orientations of images (default False)") parser.add_option("--pairwiseccc", type="string", default= None, help="Input/output pairwise ccc file") (options, args) = parser.parse_args() global_def.BATCH = True if options.dd: nargs = len(args) if nargs != 3: print "must provide name of input and two output files!" return stack = args[0] new_stack = args[1] from utilities import model_circle from statistics import ccc from statistics import mono lend = EMUtil.get_image_count(stack) lccc = [None]*(lend*(lend-1)/2) for i in xrange(lend-1): v1 = get_im( stack, i ) if( i == 0 and nargs == 2): nx = v1.get_xsize() ny = v1.get_ysize() nz = v1.get_ysize() if options.ou < 1 : radius = nx//2-2 else: radius = options.ou mask = model_circle(radius, nx, ny, nz) else: mask = get_im(args[2]) for j in xrange(i+1, lend): lccc[mono(i,j)] = [ccc(v1, get_im( stack, j ), mask), 0] order = tsp(lccc) if(len(order) != lend): print " problem with data length" from sys import exit exit() print "Total sum of cccs :",TotalDistance(order, lccc) print "ordering :",order for i in xrange(lend): get_im(stack, order[i]).write_image( new_stack, i ) elif options.align: nargs = len(args) if nargs != 3: print "must provide name of input and two output files!" return from utilities import get_params2D, model_circle from fundamentals import rot_shift2D from statistics import ccc from time import time from alignment import align2d, align2d_scf from multi_shc import mult_transform stack = args[0] new_stack = args[1] d = EMData.read_images(stack) """ # will align anyway try: ttt = d[0].get_attr('xform.params2d') for i in xrange(len(d)): alpha, sx, sy, mirror, scale = get_params2D(d[i]) d[i] = rot_shift2D(d[i], alpha, sx, sy, mirror) except: pass """ nx = d[0].get_xsize() ny = d[0].get_ysize() if options.ou < 1 : radius = nx//2-2 else: radius = options.ou mask = model_circle(radius, nx, ny) if(options.xr < 0): xrng = 0 else: xrng = options.xr if(options.yr < 0): yrng = xrng else: yrng = options.yr initial = max(options.initial, 0) from statistics import mono lend = len(d) lccc = [None]*(lend*(lend-1)/2) from utilities import read_text_row if options.pairwiseccc == None or not os.path.exists(options.pairwiseccc) : st = time() for i in xrange(lend-1): for j in xrange(i+1, lend): # j>i meaning mono entry (i,j) or (j,i) indicates T i->j (from smaller index to larger) #alpha, sx, sy, mir, peak = align2d(d[i],d[j], xrng, yrng, step=options.ts, first_ring=options.ir, last_ring=radius, mode = "F") alpha, sx, sy, mir, peak = align2d_scf(d[i],d[j], xrng, yrng, ou=radius) lccc[mono(i,j)] = [ccc(d[j], rot_shift2D(d[i], alpha, sx, sy, mir, 1.0), mask), alpha, sx, sy, mir] #print " %4d %10.1f"%(i,time()-st) if(not os.path.exists(options.pairwiseccc)): from utilities import write_text_row write_text_row([[initial,0,0,0,0]]+lccc,options.pairwiseccc) elif(os.path.exists(options.pairwiseccc)): lccc = read_text_row(options.pairwiseccc) initial = int(lccc[0][0] + 0.1) del lccc[0] for i in xrange(len(lccc)): T = Transform({"type":"2D","alpha":lccc[i][1],"tx":lccc[i][2],"ty":lccc[i][3],"mirror":int(lccc[i][4]+0.1)}) lccc[i] = [lccc[i][0],T] tdummy = Transform({"type":"2D"}) maxsum = -1.023 for m in xrange(0,lend):#initial, initial+1): indc = range( lend ) lsnake = [[m, tdummy, 0.0]] del indc[m] lsum = 0.0 while len(indc) > 1: maxcit = -111. for i in xrange(len(indc)): cuc = lccc[mono(indc[i], lsnake[-1][0])][0] if cuc > maxcit: maxcit = cuc qi = indc[i] # Here we need transformation from the current to the previous, # meaning indc[i] -> lsnake[-1][0] T = lccc[mono(indc[i], lsnake[-1][0])][1] # If direction is from larger to smaller index, the transformation has to be inverted if( indc[i] > lsnake[-1][0] ): T = T.inverse() lsnake.append([qi,T, maxcit]) lsum += maxcit del indc[indc.index(qi)] T = lccc[mono(indc[-1], lsnake[-1][0])][1] if( indc[-1] > lsnake[-1][0]): T = T.inverse() lsnake.append([indc[-1], T, lccc[mono(indc[-1], lsnake[-1][0])][0]]) print " initial image and lsum ",m,lsum #print lsnake if(lsum > maxsum): maxsum = lsum init = m snake = [lsnake[i] for i in xrange(lend)] print " Initial image selected : ",init,maxsum," ",TotalDistance([snake[m][0] for m in xrange(lend)], lccc) #for q in snake: print q from copy import deepcopy trans=deepcopy([snake[i][1] for i in xrange(len(snake))]) print [snake[i][0] for i in xrange(len(snake))] """ for m in xrange(lend): prms = trans[m].get_params("2D") print " %3d %7.1f %7.1f %7.1f %2d %6.2f"%(snake[m][0], prms["alpha"], prms["tx"], prms["ty"], prms["mirror"], snake[m][2]) """ for k in xrange(lend-2,0,-1): T = snake[k][1] for i in xrange(k+1, lend): trans[i] = T*trans[i] # To add - apply all transformations and do the overall centering. for m in xrange(lend): prms = trans[m].get_params("2D") #print " %3d %7.1f %7.1f %7.1f %2d %6.2f"%(snake[m][0], prms["alpha"], prms["tx"], prms["ty"], prms["mirror"], snake[m][2]) #rot_shift2D(d[snake[m][0]], prms["alpha"], prms["tx"], prms["ty"], prms["mirror"]).write_image(new_stack, m) rot_shift2D(d[snake[m][0]], prms["alpha"], 0.0,0.0, prms["mirror"]).write_image(new_stack, m) order = tsp(lccc) if(len(order) != lend): print " problem with data length" from sys import exit exit() print TotalDistance(order, lccc) print order ibeg = order.index(init) order = [order[(i+ibeg)%lend] for i in xrange(lend)] print TotalDistance(order, lccc) print order snake = [tdummy] for i in xrange(1,lend): # Here we need transformation from the current to the previous, # meaning order[i] -> order[i-1]] T = lccc[mono(order[i], order[i-1])][1] # If direction is from larger to smaller index, the transformation has to be inverted if( order[i] > order[i-1] ): T = T.inverse() snake.append(T) assert(len(snake) == lend) from copy import deepcopy trans = deepcopy(snake) for k in xrange(lend-2,0,-1): T = snake[k] for i in xrange(k+1, lend): trans[i] = T*trans[i] # Try to smooth the angles - complicated, I am afraid one would have to use angles forward and backwards # and find their average?? # In addition, one would have to recenter them """ trms = [] for m in xrange(lend): prms = trans[m].get_params("2D") trms.append([prms["alpha"], prms["mirror"]]) for i in xrange(3): for m in xrange(lend): mb = (m-1)%lend me = (m+1)%lend # angles order mb,m,me # calculate predicted angles mb->m """ for m in xrange(lend): prms = trans[m].get_params("2D") #rot_shift2D(d[order[m]], prms["alpha"], prms["tx"], prms["ty"], prms["mirror"]).write_image("metro.hdf", m) rot_shift2D(d[order[m]], prms["alpha"], 0.0,0.0, prms["mirror"]).write_image(args[2], m) """ # This was an effort to get number of loops, inconclusive, to say the least from numpy import outer, zeros, float32, sqrt lend = len(d) cor = zeros(lend,float32) cor = outer(cor, cor) for i in xrange(lend): cor[i][i] = 1.0 for i in xrange(lend-1): for j in xrange(i+1, lend): cor[i,j] = lccc[mono(i,j)][0] cor[j,i] = cor[i,j] lmbd, eigvec = pca(cor) from utilities import write_text_file nvec=20 print [lmbd[j] for j in xrange(nvec)] print " G" mm = [-1]*lend for i in xrange(lend): # row mi = -1.0e23 for j in xrange(nvec): qt = eigvec[j][i] if(abs(qt)>mi): mi = abs(qt) mm[i] = j for j in xrange(nvec): qt = eigvec[j][i] print round(qt,3), # eigenvector print mm[i] print for j in xrange(nvec): qt = [] for i in xrange(lend): if(mm[i] == j): qt.append(i) if(len(qt)>0): write_text_file(qt,"loop%02d.txt"%j) """ """ print [lmbd[j] for j in xrange(nvec)] print " B" mm = [-1]*lend for i in xrange(lend): # row mi = -1.0e23 for j in xrange(nvec): qt = eigvec[j][i]/sqrt(lmbd[j]) if(abs(qt)>mi): mi = abs(qt) mm[i] = j for j in xrange(nvec): qt = eigvec[j][i]/sqrt(lmbd[j]) print round(qt,3), # eigenvector print mm[i] print """ """ lend=3 cor = zeros(lend,float32) cor = outer(cor, cor) cor[0][0] =136.77 cor[0][1] = 79.15 cor[0][2] = 37.13 cor[1][0] = 79.15 cor[2][0] = 37.13 cor[1][1] = 50.04 cor[1][2] = 21.65 cor[2][1] = 21.65 cor[2][2] = 13.26 lmbd, eigvec = pca(cor) print lmbd print eigvec for i in xrange(lend): # row for j in xrange(lend): print eigvec[j][i], # eigenvector print print " B" for i in xrange(lend): # row for j in xrange(lend): print eigvec[j][i]/sqrt(lmbd[j]), # eigenvector print print " G" for i in xrange(lend): # row for j in xrange(lend): print eigvec[j][i]*sqrt(lmbd[j]), # eigenvector print """ else: nargs = len(args) if nargs != 2: print "must provide name of input and output file!" return from utilities import get_params2D, model_circle from fundamentals import rot_shift2D from statistics import ccc from time import time from alignment import align2d from multi_shc import mult_transform stack = args[0] new_stack = args[1] d = EMData.read_images(stack) try: ttt = d[0].get_attr('xform.params2d') for i in xrange(len(d)): alpha, sx, sy, mirror, scale = get_params2D(d[i]) d[i] = rot_shift2D(d[i], alpha, sx, sy, mirror) except: pass nx = d[0].get_xsize() ny = d[0].get_ysize() if options.radius < 1 : radius = nx//2-2 else: radius = options.radius mask = model_circle(radius, nx, ny) init = options.initial if init > -1 : print " initial image: %d" % init temp = d[init].copy() temp.write_image(new_stack, 0) del d[init] k = 1 lsum = 0.0 while len(d) > 1: maxcit = -111. for i in xrange(len(d)): cuc = ccc(d[i], temp, mask) if cuc > maxcit: maxcit = cuc qi = i # print k, maxcit lsum += maxcit temp = d[qi].copy() del d[qi] temp.write_image(new_stack, k) k += 1 print lsum d[0].write_image(new_stack, k) else: if options.circular : print "Using options.circular" # figure the "best circular" starting image maxsum = -1.023 for m in xrange(len(d)): indc = range(len(d) ) lsnake = [-1]*(len(d)+1) lsnake[0] = m lsnake[-1] = m del indc[m] temp = d[m].copy() lsum = 0.0 direction = +1 k = 1 while len(indc) > 1: maxcit = -111. for i in xrange(len(indc)): cuc = ccc(d[indc[i]], temp, mask) if cuc > maxcit: maxcit = cuc qi = indc[i] lsnake[k] = qi lsum += maxcit del indc[indc.index(qi)] direction = -direction for i in xrange( 1,len(d) ): if( direction > 0 ): if(lsnake[i] == -1): temp = d[lsnake[i-1]].copy() #print " forw ",lsnake[i-1] k = i break else: if(lsnake[len(d) - i] == -1): temp = d[lsnake[len(d) - i +1]].copy() #print " back ",lsnake[len(d) - i +1] k = len(d) - i break lsnake[lsnake.index(-1)] = indc[-1] #print " initial image and lsum ",m,lsum #print lsnake if(lsum > maxsum): maxsum = lsum init = m snake = [lsnake[i] for i in xrange(len(d))] print " Initial image selected : ",init,maxsum print lsnake for m in xrange(len(d)): d[snake[m]].write_image(new_stack, m) else: # figure the "best" starting image maxsum = -1.023 for m in xrange(len(d)): indc = range(len(d) ) lsnake = [m] del indc[m] temp = d[m].copy() lsum = 0.0 while len(indc) > 1: maxcit = -111. for i in xrange(len(indc)): cuc = ccc(d[indc[i]], temp, mask) if cuc > maxcit: maxcit = cuc qi = indc[i] lsnake.append(qi) lsum += maxcit temp = d[qi].copy() del indc[indc.index(qi)] lsnake.append(indc[-1]) #print " initial image and lsum ",m,lsum #print lsnake if(lsum > maxsum): maxsum = lsum init = m snake = [lsnake[i] for i in xrange(len(d))] print " Initial image selected : ",init,maxsum print lsnake for m in xrange(len(d)): d[snake[m]].write_image(new_stack, m)
def main(): from optparse import OptionParser from global_def import SPARXVERSION from EMAN2 import EMData from logger import Logger, BaseLogger_Files import sys, os, time global Tracker, Blockdata from global_def import ERROR progname = os.path.basename(sys.argv[0]) usage = progname + " --output_dir=output_dir --isac_dir=output_dir_of_isac " parser = OptionParser(usage, version=SPARXVERSION) parser.add_option( "--adjust_to_analytic_model", action="store_true", default=False, help="adjust power spectrum of 2-D averages to an analytic model ") parser.add_option( "--adjust_to_given_pw2", action="store_true", default=False, help="adjust power spectrum to 2-D averages to given 1D power spectrum" ) parser.add_option("--B_enhance", action="store_true", default=False, help="using B-factor to enhance 2-D averages") parser.add_option("--no_adjustment", action="store_true", default=False, help="No power spectrum adjustment") options_list = [] adjust_to_analytic_model = False for q in sys.argv[1:]: if (q[:26] == "--adjust_to_analytic_model"): adjust_to_analytic_model = True options_list.append(q) break adjust_to_given_pw2 = False for q in sys.argv[1:]: if (q[:21] == "--adjust_to_given_pw2"): adjust_to_given_pw2 = True options_list.append(q) break B_enhance = False for q in sys.argv[1:]: if (q[:11] == "--B_enhance"): B_enhance = True options_list.append(q) break no_adjustment = False for q in sys.argv[1:]: if (q[:15] == "--no_adjustment"): no_adjustment = True options_list.append(q) break if len(options_list) == 0: if (Blockdata["myid"] == Blockdata["main_node"]): print( "specify one of the following options to start: 1. adjust_to_analytic_model; 2. adjust_to_given_pw2; 3. B_enhance; 4. no_adjustment" ) if len(options_list) > 1: ERROR( "The specified options are exclusive. Use only one of them to start", "sxcompute_isac_avg.py", 1, Blockdata["myid"]) # options in common parser.add_option( "--isac_dir", type="string", default='', help="ISAC run output directory, input directory for this command") parser.add_option( "--output_dir", type="string", default='', help="output directory where computed averages are saved") parser.add_option("--pixel_size", type="float", default=-1.0, help="pixel_size of raw images") parser.add_option( "--fl", type="float", default=-1.0, help= "low pass filter, =-1, not applied; =1, using FH1 (initial resolution), =2 using FH2 (resolution after local alignment), or user provided value" ) parser.add_option("--stack", type="string", default="", help="data stack used in ISAC") parser.add_option("--radius", type="int", default=-1, help="radius") parser.add_option("--xr", type="float", default=-1.0, help="local alignment search range") parser.add_option("--ts", type="float", default=1.0, help="local alignment search step") parser.add_option("--fh", type="float", default=-1., help="local alignment high frequencies limit") parser.add_option("--maxit", type="int", default=5, help="local alignment iterations") parser.add_option("--navg", type="int", default=-1, help="number of aveages") parser.add_option("--skip_local_alignment", action="store_true", default=False, help="skip local alignment") parser.add_option( "--noctf", action="store_true", default=False, help= "no ctf correction, useful for negative stained data. always ctf for cryo data" ) if B_enhance: parser.add_option( "--B_start", type="float", default=10.0, help= "start frequency (1./Angstrom) of power spectrum for B_factor estimation" ) parser.add_option( "--Bfactor", type="float", default=-1.0, help= "User defined bactors (e.g. 45.0[A^2]). By default, the program automatically estimates B-factor. " ) if adjust_to_given_pw2: parser.add_option("--modelpw", type="string", default='', help="1-D reference power spectrum") checking_flag = 0 if (Blockdata["myid"] == Blockdata["main_node"]): if not os.path.exists(options.modelpw): checking_flag = 1 checking_flag = bcast_number_to_all(checking_flag, Blockdata["main_node"], MPI_COMM_WORLD) if checking_flag == 1: ERROR("User provided power spectrum does not exist", "sxcompute_isac_avg.py", 1, Blockdata["myid"]) (options, args) = parser.parse_args(sys.argv[1:]) Tracker = {} Constants = {} Constants["isac_dir"] = options.isac_dir Constants["masterdir"] = options.output_dir Constants["pixel_size"] = options.pixel_size Constants["orgstack"] = options.stack Constants["radius"] = options.radius Constants["xrange"] = options.xr Constants["xstep"] = options.ts Constants["FH"] = options.fh Constants["maxit"] = options.maxit Constants["navg"] = options.navg Constants["low_pass_filter"] = options.fl if B_enhance: Constants["B_start"] = options.B_start Constants["Bfactor"] = options.Bfactor if adjust_to_given_pw2: Constants["modelpw"] = options.modelpw Tracker["constants"] = Constants # ------------------------------------------------------------- # # Create and initialize Tracker dictionary with input options # State Variables #<<<---------------------->>>imported functions<<<--------------------------------------------- from utilities import get_im, bcast_number_to_all, write_text_file, read_text_file, wrap_mpi_bcast, write_text_row from utilities import cmdexecute from filter import filt_tanl from time import sleep from logger import Logger, BaseLogger_Files import user_functions import string from string import split, atoi, atof import json #x_range = max(Tracker["constants"]["xrange"], int(1./Tracker["ini_shrink"])+1) #y_range = x_range ####----------------------------------------------------------- # Create Master directory line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" if Tracker["constants"]["masterdir"] == Tracker["constants"]["isac_dir"]: masterdir = os.path.join(Tracker["constants"]["isac_dir"], "sharpen") else: masterdir = Tracker["constants"]["masterdir"] if (Blockdata["myid"] == Blockdata["main_node"]): msg = "Postprocessing ISAC 2D averages starts" print(line, "Postprocessing ISAC 2D averages starts") if not masterdir: timestring = strftime("_%d_%b_%Y_%H_%M_%S", localtime()) masterdir = "sharpen_" + Tracker["constants"]["isac_dir"] os.mkdir(masterdir) else: if os.path.exists(masterdir): print("%s already exists" % masterdir) else: os.mkdir(masterdir) li = len(masterdir) else: li = 0 li = mpi_bcast(li, 1, MPI_INT, Blockdata["main_node"], MPI_COMM_WORLD)[0] masterdir = mpi_bcast(masterdir, li, MPI_CHAR, Blockdata["main_node"], MPI_COMM_WORLD) masterdir = string.join(masterdir, "") Tracker["constants"]["masterdir"] = masterdir log_main = Logger(BaseLogger_Files()) log_main.prefix = Tracker["constants"]["masterdir"] + "/" while not os.path.exists(Tracker["constants"]["masterdir"]): print("Node ", Blockdata["myid"], " waiting...", Tracker["constants"]["masterdir"]) sleep(1) mpi_barrier(MPI_COMM_WORLD) if (Blockdata["myid"] == Blockdata["main_node"]): init_dict = {} print(Tracker["constants"]["isac_dir"]) Tracker["directory"] = os.path.join(Tracker["constants"]["isac_dir"], "2dalignment") core = read_text_row( os.path.join(Tracker["directory"], "initial2Dparams.txt")) for im in xrange(len(core)): init_dict[im] = core[im] del core else: init_dict = 0 init_dict = wrap_mpi_bcast(init_dict, Blockdata["main_node"], communicator=MPI_COMM_WORLD) ### if (Blockdata["myid"] == Blockdata["main_node"]): #Tracker["constants"]["orgstack"] = "bdb:"+ os.path.join(Tracker["constants"]["isac_dir"],"../","sparx_stack") image = get_im(Tracker["constants"]["orgstack"], 0) Tracker["constants"]["nnxo"] = image.get_xsize() try: ctf_params = image.get_attr("ctf") if Tracker["constants"]["pixel_size"] == -1.: Tracker["constants"]["pixel_size"] = ctf_params.apix except: print("pixel size value is not given.") Tracker["ini_shrink"] = float( get_im(os.path.join(Tracker["directory"], "aqfinal.hdf"), 0).get_xsize()) / Tracker["constants"]["nnxo"] else: Tracker["ini_shrink"] = 0 Tracker = wrap_mpi_bcast(Tracker, Blockdata["main_node"], communicator=MPI_COMM_WORLD) #print(Tracker["constants"]["pixel_size"], "pixel_size") x_range = max(Tracker["constants"]["xrange"], int(1. / Tracker["ini_shrink"]) + 1) y_range = x_range if (Blockdata["myid"] == Blockdata["main_node"]): parameters = read_text_row( os.path.join(Tracker["constants"]["isac_dir"], "all_parameters.txt")) else: parameters = 0 parameters = wrap_mpi_bcast(parameters, Blockdata["main_node"], communicator=MPI_COMM_WORLD) params_dict = {} list_dict = {} #parepare params_dict if Tracker["constants"]["navg"] < 0: navg = EMUtil.get_image_count( os.path.join(Tracker["constants"]["isac_dir"], "class_averages.hdf")) else: navg = min( Tracker["constants"]["navg"], EMUtil.get_image_count( os.path.join(Tracker["constants"]["isac_dir"], "class_averages.hdf"))) global_dict = {} ptl_list = [] memlist = [] if (Blockdata["myid"] == Blockdata["main_node"]): for iavg in xrange(navg): params_of_this_average = [] image = get_im( os.path.join(Tracker["constants"]["isac_dir"], "class_averages.hdf"), iavg) members = image.get_attr("members") memlist.append(members) for im in xrange(len(members)): abs_id = members[im] global_dict[abs_id] = [iavg, im] P = combine_params2( init_dict[abs_id][0], init_dict[abs_id][1], init_dict[abs_id][2], init_dict[abs_id][3], \ parameters[abs_id][0], parameters[abs_id][1]/Tracker["ini_shrink"], parameters[abs_id][2]/Tracker["ini_shrink"], parameters[abs_id][3]) if parameters[abs_id][3] == -1: print("wrong one") params_of_this_average.append([P[0], P[1], P[2], P[3], 1.0]) ptl_list.append(abs_id) params_dict[iavg] = params_of_this_average list_dict[iavg] = members write_text_row( params_of_this_average, os.path.join(Tracker["constants"]["masterdir"], "params_avg_%03d.txt" % iavg)) ptl_list.sort() init_params = [None for im in xrange(len(ptl_list))] for im in xrange(len(ptl_list)): init_params[im] = [ptl_list[im]] + params_dict[global_dict[ ptl_list[im]][0]][global_dict[ptl_list[im]][1]] write_text_row( init_params, os.path.join(Tracker["constants"]["masterdir"], "init_isac_params.txt")) else: params_dict = 0 list_dict = 0 memlist = 0 params_dict = wrap_mpi_bcast(params_dict, Blockdata["main_node"], communicator=MPI_COMM_WORLD) list_dict = wrap_mpi_bcast(list_dict, Blockdata["main_node"], communicator=MPI_COMM_WORLD) memlist = wrap_mpi_bcast(memlist, Blockdata["main_node"], communicator=MPI_COMM_WORLD) # Now computing! del init_dict tag_sharpen_avg = 1000 ## always apply low pass filter to B_enhanced images to suppress noise in high frequencies enforced_to_H1 = False if options.B_enhance: if Tracker["constants"]["low_pass_filter"] == -1: print("User does not provide low pass filter") enforced_to_H1 = True if navg < Blockdata["nproc"]: # Each CPU do one average FH_list = [None for im in xrange(navg)] for iavg in xrange(navg): if Blockdata["myid"] == iavg: mlist = [None for i in xrange(len(list_dict[iavg]))] for im in xrange(len(mlist)): mlist[im] = get_im(Tracker["constants"]["orgstack"], list_dict[iavg][im]) set_params2D(mlist[im], params_dict[iavg][im], xform="xform.align2d") if options.noctf: new_avg, frc, plist = compute_average_noctf( mlist, Tracker["constants"]["radius"]) else: new_avg, frc, plist = compute_average_ctf( mlist, Tracker["constants"]["radius"]) FH1 = get_optimistic_res(frc) #write_text_file(frc, os.path.join(Tracker["constants"]["masterdir"], "fsc%03d_before_ali.txt"%iavg)) if not options.skip_local_alignment: new_average1 = within_group_refinement([mlist[kik] for kik in xrange(0,len(mlist),2)], maskfile= None, randomize= False, ir=1.0, \ ou=Tracker["constants"]["radius"], rs=1.0, xrng=[x_range], yrng=[y_range], step=[Tracker["constants"]["xstep"]], \ dst=0.0, maxit=Tracker["constants"]["maxit"], FH = max(Tracker["constants"]["FH"], FH1), FF=0.1) new_average2 = within_group_refinement([mlist[kik] for kik in xrange(1,len(mlist),2)], maskfile= None, randomize= False, ir=1.0, \ ou=Tracker["constants"]["radius"], rs=1.0, xrng=[x_range], yrng=[y_range], step=[Tracker["constants"]["xstep"]], \ dst=0.0, maxit=Tracker["constants"]["maxit"], FH = max(Tracker["constants"]["FH"], FH1), FF=0.1) if options.noctf: new_avg, frc, plist = compute_average_noctf( mlist, Tracker["constants"]["radius"]) else: new_avg, frc, plist = compute_average_ctf( mlist, Tracker["constants"]["radius"]) FH2 = get_optimistic_res(frc) #write_text_file(frc, os.path.join(Tracker["constants"]["masterdir"], "fsc%03d.txt"%iavg)) #if Tracker["constants"]["nopwadj"]: # pw adjustment, 1. analytic model 2. PDB model 3. B-facttor enhancement else: FH2 = 0.0 FH_list[iavg] = [FH1, FH2] if options.B_enhance: new_avg, gb = apply_enhancement( new_avg, Tracker["constants"]["B_start"], Tracker["constants"]["pixel_size"], Tracker["constants"]["Bfactor"]) print("Process avg %d %f %f %f" % (iavg, gb, FH1, FH2)) elif options.adjust_to_given_pw2: roo = read_text_file(Tracker["constants"]["modelpw"], -1) roo = roo[0] # always put pw in the first column new_avg = adjust_pw_to_model( new_avg, Tracker["constants"]["pixel_size"], roo) elif options.adjust_to_analytic_model: new_avg = adjust_pw_to_model( new_avg, Tracker["constants"]["pixel_size"], None) elif options.no_adjustment: pass print("Process avg %d %f %f" % (iavg, FH1, FH2)) if Tracker["constants"]["low_pass_filter"] != -1.: if Tracker["constants"]["low_pass_filter"] == 1.: low_pass_filter = FH1 elif Tracker["constants"]["low_pass_filter"] == 2.: low_pass_filter = FH2 if options.skip_local_alignment: low_pass_filter = FH1 else: low_pass_filter = Tracker["constants"][ "low_pass_filter"] if low_pass_filter >= 0.45: low_pass_filter = 0.45 new_avg = filt_tanl(new_avg, low_pass_filter, 0.1) new_avg.set_attr("members", list_dict[iavg]) new_avg.set_attr("n_objects", len(list_dict[iavg])) mpi_barrier(MPI_COMM_WORLD) for im in xrange(navg): # avg if im == Blockdata[ "myid"] and Blockdata["myid"] != Blockdata["main_node"]: send_EMData(new_avg, Blockdata["main_node"], tag_sharpen_avg) elif Blockdata["myid"] == Blockdata["main_node"]: if im != Blockdata["main_node"]: new_avg_other_cpu = recv_EMData(im, tag_sharpen_avg) new_avg_other_cpu.set_attr("members", memlist[im]) new_avg_other_cpu.write_image( os.path.join(Tracker["constants"]["masterdir"], "class_averages.hdf"), im) else: new_avg.write_image( os.path.join(Tracker["constants"]["masterdir"], "class_averages.hdf"), im) if not options.skip_local_alignment: if im == Blockdata["myid"]: write_text_row( plist, os.path.join(Tracker["constants"]["masterdir"], "ali2d_local_params_avg_%03d.txt" % im)) if Blockdata["myid"] == im and Blockdata["myid"] != Blockdata[ "main_node"]: wrap_mpi_send(plist_dict[im], Blockdata["main_node"], MPI_COMM_WORLD) elif im != Blockdata["main_node"] and Blockdata[ "myid"] == Blockdata["main_node"]: dummy = wrap_mpi_recv(im, MPI_COMM_WORLD) plist_dict[im] = dummy if im == Blockdata["myid"] and im != Blockdata["main_node"]: wrap_mpi_send(FH_list[im], Blockdata["main_node"], MPI_COMM_WORLD) elif im != Blockdata["main_node"] and Blockdata[ "myid"] == Blockdata["main_node"]: dummy = wrap_mpi_recv(im, MPI_COMM_WORLD) FH_list[im] = dummy else: if im == Blockdata["myid"] and im != Blockdata["main_node"]: wrap_mpi_send(FH_list, Blockdata["main_node"], MPI_COMM_WORLD) elif im != Blockdata["main_node"] and Blockdata[ "myid"] == Blockdata["main_node"]: dummy = wrap_mpi_recv(im, MPI_COMM_WORLD) FH_list[im] = dummy[im] mpi_barrier(MPI_COMM_WORLD) else: FH_list = [[0, 0.0, 0.0] for im in xrange(navg)] image_start, image_end = MPI_start_end(navg, Blockdata["nproc"], Blockdata["myid"]) if Blockdata["myid"] == Blockdata["main_node"]: cpu_dict = {} for iproc in xrange(Blockdata["nproc"]): local_image_start, local_image_end = MPI_start_end( navg, Blockdata["nproc"], iproc) for im in xrange(local_image_start, local_image_end): cpu_dict[im] = iproc else: cpu_dict = 0 cpu_dict = wrap_mpi_bcast(cpu_dict, Blockdata["main_node"], communicator=MPI_COMM_WORLD) slist = [None for im in xrange(navg)] ini_list = [None for im in xrange(navg)] avg1_list = [None for im in xrange(navg)] avg2_list = [None for im in xrange(navg)] plist_dict = {} data_list = [None for im in xrange(navg)] if Blockdata["myid"] == Blockdata["main_node"]: print("read data") for iavg in xrange(image_start, image_end): mlist = [None for i in xrange(len(list_dict[iavg]))] for im in xrange(len(mlist)): mlist[im] = get_im(Tracker["constants"]["orgstack"], list_dict[iavg][im]) set_params2D(mlist[im], params_dict[iavg][im], xform="xform.align2d") data_list[iavg] = mlist print("read data done %d" % Blockdata["myid"]) #if Blockdata["myid"] == Blockdata["main_node"]: print("start to compute averages") for iavg in xrange(image_start, image_end): mlist = data_list[iavg] if options.noctf: new_avg, frc, plist = compute_average_noctf( mlist, Tracker["constants"]["radius"]) else: new_avg, frc, plist = compute_average_ctf( mlist, Tracker["constants"]["radius"]) FH1 = get_optimistic_res(frc) #write_text_file(frc, os.path.join(Tracker["constants"]["masterdir"], "fsc%03d_before_ali.txt"%iavg)) if not options.skip_local_alignment: new_average1 = within_group_refinement([mlist[kik] for kik in xrange(0,len(mlist),2)], maskfile= None, randomize= False, ir=1.0, \ ou=Tracker["constants"]["radius"], rs=1.0, xrng=[x_range], yrng=[y_range], step=[Tracker["constants"]["xstep"]], \ dst=0.0, maxit=Tracker["constants"]["maxit"], FH=max(Tracker["constants"]["FH"], FH1), FF=0.1) new_average2 = within_group_refinement([mlist[kik] for kik in xrange(1,len(mlist),2)], maskfile= None, randomize= False, ir=1.0, \ ou= Tracker["constants"]["radius"], rs=1.0, xrng=[ x_range], yrng=[y_range], step=[Tracker["constants"]["xstep"]], \ dst=0.0, maxit=Tracker["constants"]["maxit"], FH = max(Tracker["constants"]["FH"], FH1), FF=0.1) if options.noctf: new_avg, frc, plist = compute_average_noctf( mlist, Tracker["constants"]["radius"]) else: new_avg, frc, plist = compute_average_ctf( mlist, Tracker["constants"]["radius"]) plist_dict[iavg] = plist FH2 = get_optimistic_res(frc) else: FH2 = 0.0 #write_text_file(frc, os.path.join(Tracker["constants"]["masterdir"], "fsc%03d.txt"%iavg)) FH_list[iavg] = [iavg, FH1, FH2] if options.B_enhance: new_avg, gb = apply_enhancement( new_avg, Tracker["constants"]["B_start"], Tracker["constants"]["pixel_size"], Tracker["constants"]["Bfactor"]) print("Process avg %d %f %f %f" % (iavg, gb, FH1, FH2)) elif options.adjust_to_given_pw2: roo = read_text_file(Tracker["constants"]["modelpw"], -1) roo = roo[0] # always on the first column new_avg = adjust_pw_to_model( new_avg, Tracker["constants"]["pixel_size"], roo) print("Process avg %d %f %f" % (iavg, FH1, FH2)) elif adjust_to_analytic_model: new_avg = adjust_pw_to_model( new_avg, Tracker["constants"]["pixel_size"], None) print("Process avg %d %f %f" % (iavg, FH1, FH2)) elif options.no_adjustment: pass if Tracker["constants"]["low_pass_filter"] != -1.: new_avg = filt_tanl(new_avg, Tracker["constants"]["low_pass_filter"], 0.1) if Tracker["constants"]["low_pass_filter"] != -1.: if Tracker["constants"]["low_pass_filter"] == 1.: low_pass_filter = FH1 elif Tracker["constants"]["low_pass_filter"] == 2.: low_pass_filter = FH2 if options.skip_local_alignment: low_pass_filter = FH1 else: low_pass_filter = Tracker["constants"]["low_pass_filter"] if low_pass_filter >= 0.45: low_pass_filter = 0.45 new_avg = filt_tanl(new_avg, low_pass_filter, 0.1) else: if enforced_to_H1: new_avg = filt_tanl(new_avg, FH1, 0.1) if options.B_enhance: new_avg = fft(new_avg) new_avg.set_attr("members", list_dict[iavg]) new_avg.set_attr("n_objects", len(list_dict[iavg])) slist[iavg] = new_avg ## send to main node to write mpi_barrier(MPI_COMM_WORLD) for im in xrange(navg): # avg if cpu_dict[im] == Blockdata[ "myid"] and Blockdata["myid"] != Blockdata["main_node"]: send_EMData(slist[im], Blockdata["main_node"], tag_sharpen_avg) elif cpu_dict[im] == Blockdata["myid"] and Blockdata[ "myid"] == Blockdata["main_node"]: slist[im].set_attr("members", memlist[im]) slist[im].write_image( os.path.join(Tracker["constants"]["masterdir"], "class_averages.hdf"), im) elif cpu_dict[im] != Blockdata["myid"] and Blockdata[ "myid"] == Blockdata["main_node"]: new_avg_other_cpu = recv_EMData(cpu_dict[im], tag_sharpen_avg) new_avg_other_cpu.set_attr("members", memlist[im]) new_avg_other_cpu.write_image( os.path.join(Tracker["constants"]["masterdir"], "class_averages.hdf"), im) if not options.skip_local_alignment: if cpu_dict[im] == Blockdata["myid"]: write_text_row( plist_dict[im], os.path.join(Tracker["constants"]["masterdir"], "ali2d_local_params_avg_%03d.txt" % im)) if cpu_dict[im] == Blockdata[ "myid"] and cpu_dict[im] != Blockdata["main_node"]: wrap_mpi_send(plist_dict[im], Blockdata["main_node"], MPI_COMM_WORLD) wrap_mpi_send(FH_list, Blockdata["main_node"], MPI_COMM_WORLD) elif cpu_dict[im] != Blockdata["main_node"] and Blockdata[ "myid"] == Blockdata["main_node"]: dummy = wrap_mpi_recv(cpu_dict[im], MPI_COMM_WORLD) plist_dict[im] = dummy dummy = wrap_mpi_recv(cpu_dict[im], MPI_COMM_WORLD) FH_list[im] = dummy[im] else: if cpu_dict[im] == Blockdata[ "myid"] and cpu_dict[im] != Blockdata["main_node"]: wrap_mpi_send(FH_list, Blockdata["main_node"], MPI_COMM_WORLD) elif cpu_dict[im] != Blockdata["main_node"] and Blockdata[ "myid"] == Blockdata["main_node"]: dummy = wrap_mpi_recv(cpu_dict[im], MPI_COMM_WORLD) FH_list[im] = dummy[im] mpi_barrier(MPI_COMM_WORLD) mpi_barrier(MPI_COMM_WORLD) if not options.skip_local_alignment: if Blockdata["myid"] == Blockdata["main_node"]: ali3d_local_params = [None for im in xrange(len(ptl_list))] for im in xrange(len(ptl_list)): ali3d_local_params[im] = [ptl_list[im]] + plist_dict[ global_dict[ptl_list[im]][0]][global_dict[ptl_list[im]][1]] write_text_row( ali3d_local_params, os.path.join(Tracker["constants"]["masterdir"], "ali2d_local_params.txt")) write_text_row( FH_list, os.path.join(Tracker["constants"]["masterdir"], "FH_list.txt")) else: if Blockdata["myid"] == Blockdata["main_node"]: write_text_row( FH_list, os.path.join(Tracker["constants"]["masterdir"], "FH_list.txt")) mpi_barrier(MPI_COMM_WORLD) target_xr = 3 target_yr = 3 if (Blockdata["myid"] == 0): cmd = "{} {} {} {} {} {} {} {} {} {}".format("sxchains.py", os.path.join(Tracker["constants"]["masterdir"],"class_averages.hdf"),\ os.path.join(Tracker["constants"]["masterdir"],"junk.hdf"),os.path.join(Tracker["constants"]["masterdir"],"ordered_class_averages.hdf"),\ "--circular","--radius=%d"%Tracker["constants"]["radius"] , "--xr=%d"%(target_xr+1),"--yr=%d"%(target_yr+1),"--align", ">/dev/null") junk = cmdexecute(cmd) cmd = "{} {}".format( "rm -rf", os.path.join(Tracker["constants"]["masterdir"], "junk.hdf")) junk = cmdexecute(cmd) from mpi import mpi_finalize mpi_finalize() exit()
def main(): import sys import os import math import random import pyemtbx.options import time from random import random, seed, randint from optparse import OptionParser progname = os.path.basename(sys.argv[0]) usage = progname + """ [options] <inputfile> <outputfile> Generic 2-D image processing programs. Functionality: 1. Phase flip a stack of images and write output to new file: sxprocess.py input_stack.hdf output_stack.hdf --phase_flip 2. Resample (decimate or interpolate up) images (2D or 3D) in a stack to change the pixel size. The window size will change accordingly. sxprocess input.hdf output.hdf --changesize --ratio=0.5 3. Compute average power spectrum of a stack of 2D images with optional padding (option wn) with zeroes. sxprocess.py input_stack.hdf powerspectrum.hdf --pw [--wn=1024] 4. Generate a stack of projections bdb:data and micrographs with prefix mic (i.e., mic0.hdf, mic1.hdf etc) from structure input_structure.hdf, with CTF applied to both projections and micrographs: sxprocess.py input_structure.hdf data mic --generate_projections format="bdb":apix=5.2:CTF=True:boxsize=64 5. Retrieve original image numbers in the selected ISAC group (here group 12 from generation 3): sxprocess.py bdb:test3 class_averages_generation_3.hdf list3_12.txt --isacgroup=12 --params=originalid 6. Retrieve original image numbers of images listed in ISAC output stack of averages: sxprocess.py select1.hdf ohk.txt 7. Adjust rotationally averaged power spectrum of an image to that of a reference image or a reference 1D power spectrum stored in an ASCII file. Optionally use a tangent low-pass filter. Also works for a stack of images, in which case the output is also a stack. sxprocess.py vol.hdf ref.hdf avol.hdf < 0.25 0.2> --adjpw sxprocess.py vol.hdf pw.txt avol.hdf < 0.25 0.2> --adjpw 8. Generate a 1D rotationally averaged power spectrum of an image. sxprocess.py vol.hdf --rotwp=rotpw.txt # Output will contain three columns: (1) rotationally averaged power spectrum (2) logarithm of the rotationally averaged power spectrum (3) integer line number (from zero to approximately to half the image size) 9. Apply 3D transformation (rotation and/or shift) to a set of orientation parameters associated with projection data. sxprocess.py --transfromparams=phi,theta,psi,tx,ty,tz input.txt output.txt The output file is then imported and 3D transformed volume computed: sxheader.py bdb:p --params=xform.projection --import=output.txt mpirun -np 2 sxrecons3d_n.py bdb:p tvol.hdf --MPI The reconstructed volume is in the position of the volume computed using the input.txt parameters and then transformed with rot_shift3D(vol, phi,theta,psi,tx,ty,tz) 10. Import ctf parameters from the output of sxcter into windowed particle headers. There are three possible input files formats: (1) all particles are in one stack, (2 aor 3) particles are in stacks, each stack corresponds to a single micrograph. In each case the particles should contain a name of the micrograph of origin stores using attribute name 'ptcl_source_image'. Normally this is done by e2boxer.py during windowing. Particles whose defocus or astigmatism error exceed set thresholds will be skipped, otherwise, virtual stacks with the original way preceded by G will be created. sxprocess.py --input=bdb:data --importctf=outdir/partres --defocuserror=10.0 --astigmatismerror=5.0 # Output will be a vritual stack bdb:Gdata sxprocess.py --input="bdb:directory/stacks*" --importctf=outdir/partres --defocuserror=10.0 --astigmatismerror=5.0 To concatenate output files: cd directory e2bdb.py . --makevstack=bdb:allparticles --filt=G IMPORTANT: Please do not move (or remove!) any input/intermediate EMAN2DB files as the information is linked between them. 11. Scale 3D shifts. The shifts in the input five columns text file with 3D orientation parameters will be DIVIDED by the scale factor sxprocess.py orientationparams.txt scaledparams.txt scale=0.5 12. Generate adaptive mask from a given 3-D volume. """ parser = OptionParser(usage, version=SPARXVERSION) parser.add_option( "--order", action="store_true", help= "Two arguments are required: name of input stack and desired name of output stack. The output stack is the input stack sorted by similarity in terms of cross-correlation coefficent.", default=False) parser.add_option("--order_lookup", action="store_true", help="Test/Debug.", default=False) parser.add_option("--order_metropolis", action="store_true", help="Test/Debug.", default=False) parser.add_option("--order_pca", action="store_true", help="Test/Debug.", default=False) parser.add_option( "--initial", type="int", default=-1, help= "Specifies which image will be used as an initial seed to form the chain. (default = 0, means the first image)" ) parser.add_option( "--circular", action="store_true", help= "Select circular ordering (fisr image has to be similar to the last", default=False) parser.add_option( "--radius", type="int", default=-1, help="Radius of a circular mask for similarity based ordering") parser.add_option( "--changesize", action="store_true", help= "resample (decimate or interpolate up) images (2D or 3D) in a stack to change the pixel size.", default=False) parser.add_option( "--ratio", type="float", default=1.0, help= "The ratio of new to old image size (if <1 the pixel size will increase and image size decrease, if>1, the other way round" ) parser.add_option( "--pw", action="store_true", help= "compute average power spectrum of a stack of 2-D images with optional padding (option wn) with zeroes", default=False) parser.add_option( "--wn", type="int", default=-1, help= "Size of window to use (should be larger/equal than particle box size, default padding to max(nx,ny))" ) parser.add_option("--phase_flip", action="store_true", help="Phase flip the input stack", default=False) parser.add_option( "--makedb", metavar="param1=value1:param2=value2", type="string", action="append", help= "One argument is required: name of key with which the database will be created. Fill in database with parameters specified as follows: --makedb param1=value1:param2=value2, e.g. 'gauss_width'=1.0:'pixel_input'=5.2:'pixel_output'=5.2:'thr_low'=1.0" ) parser.add_option( "--generate_projections", metavar="param1=value1:param2=value2", type="string", action="append", help= "Three arguments are required: name of input structure from which to generate projections, desired name of output projection stack, and desired prefix for micrographs (e.g. if prefix is 'mic', then micrographs mic0.hdf, mic1.hdf etc will be generated). Optional arguments specifying format, apix, box size and whether to add CTF effects can be entered as follows after --generate_projections: format='bdb':apix=5.2:CTF=True:boxsize=100, or format='hdf', etc., where format is bdb or hdf, apix (pixel size) is a float, CTF is True or False, and boxsize denotes the dimension of the box (assumed to be a square). If an optional parameter is not specified, it will default as follows: format='bdb', apix=2.5, CTF=False, boxsize=64." ) parser.add_option( "--isacgroup", type="int", help= "Retrieve original image numbers in the selected ISAC group. See ISAC documentation for details.", default=-1) parser.add_option( "--isacselect", action="store_true", help= "Retrieve original image numbers of images listed in ISAC output stack of averages. See ISAC documentation for details.", default=False) parser.add_option( "--params", type="string", default=None, help="Name of header of parameter, which one depends on specific option" ) parser.add_option( "--adjpw", action="store_true", help="Adjust rotationally averaged power spectrum of an image", default=False) parser.add_option( "--rotpw", type="string", default=None, help= "Name of the text file to contain rotationally averaged power spectrum of the input image." ) parser.add_option( "--transformparams", type="string", default=None, help= "Transform 3D projection orientation parameters using six 3D parameters (phi, theta,psi,sx,sy,sz). Input: --transformparams=45.,66.,12.,-2,3,-5.5 desired six transformation of the reconstructed structure. Output: file with modified orientation parameters." ) # import ctf estimates done using cter parser.add_option("--input", type="string", default=None, help="Input particles.") parser.add_option( "--importctf", type="string", default=None, help="Name of the file containing CTF parameters produced by sxcter.") parser.add_option( "--defocuserror", type="float", default=1000000.0, help= "Exclude micrographs whose relative defocus error as estimated by sxcter is larger than defocuserror percent. The error is computed as (std dev defocus)/defocus*100%" ) parser.add_option( "--astigmatismerror", type="float", default=360.0, help= "Set to zero astigmatism for micrographs whose astigmatism angular error as estimated by sxcter is larger than astigmatismerror degrees." ) # import ctf estimates done using cter parser.add_option( "--scale", type="float", default=-1.0, help= "Divide shifts in the input 3D orientation parameters text file by the scale factor." ) # generate adaptive mask from an given 3-Db volue parser.add_option("--adaptive_mask", action="store_true", help="create adavptive 3-D mask from a given volume", default=False) parser.add_option( "--nsigma", type="float", default=1., help= "number of times of sigma of the input volume to obtain the the large density cluster" ) parser.add_option( "--ndilation", type="int", default=3, help= "number of times of dilation applied to the largest cluster of density" ) parser.add_option( "--kernel_size", type="int", default=11, help="convolution kernel for smoothing the edge of the mask") parser.add_option( "--gauss_standard_dev", type="int", default=9, help="stanadard deviation value to generate Gaussian edge") (options, args) = parser.parse_args() global_def.BATCH = True if options.phase_flip: nargs = len(args) if nargs != 2: print "must provide name of input and output file!" return from EMAN2 import Processor instack = args[0] outstack = args[1] nima = EMUtil.get_image_count(instack) from filter import filt_ctf for i in xrange(nima): img = EMData() img.read_image(instack, i) try: ctf = img.get_attr('ctf') except: print "no ctf information in input stack! Exiting..." return dopad = True sign = 1 binary = 1 # phase flip assert img.get_ysize() > 1 dict = ctf.to_dict() dz = dict["defocus"] cs = dict["cs"] voltage = dict["voltage"] pixel_size = dict["apix"] b_factor = dict["bfactor"] ampcont = dict["ampcont"] dza = dict["dfdiff"] azz = dict["dfang"] if dopad and not img.is_complex(): ip = 1 else: ip = 0 params = { "filter_type": Processor.fourier_filter_types.CTF_, "defocus": dz, "Cs": cs, "voltage": voltage, "Pixel_size": pixel_size, "B_factor": b_factor, "amp_contrast": ampcont, "dopad": ip, "binary": binary, "sign": sign, "dza": dza, "azz": azz } tmp = Processor.EMFourierFilter(img, params) tmp.set_attr_dict({"ctf": ctf}) tmp.write_image(outstack, i) elif options.changesize: nargs = len(args) if nargs != 2: ERROR("must provide name of input and output file!", "change size", 1) return from utilities import get_im instack = args[0] outstack = args[1] sub_rate = float(options.ratio) nima = EMUtil.get_image_count(instack) from fundamentals import resample for i in xrange(nima): resample(get_im(instack, i), sub_rate).write_image(outstack, i) elif options.isacgroup > -1: nargs = len(args) if nargs != 3: ERROR("Three files needed on input!", "isacgroup", 1) return from utilities import get_im instack = args[0] m = get_im(args[1], int(options.isacgroup)).get_attr("members") l = [] for k in m: l.append(int(get_im(args[0], k).get_attr(options.params))) from utilities import write_text_file write_text_file(l, args[2]) elif options.isacselect: nargs = len(args) if nargs != 2: ERROR("Two files needed on input!", "isacgroup", 1) return from utilities import get_im nima = EMUtil.get_image_count(args[0]) m = [] for k in xrange(nima): m += get_im(args[0], k).get_attr("members") m.sort() from utilities import write_text_file write_text_file(m, args[1]) elif options.pw: nargs = len(args) if nargs < 2: ERROR("must provide name of input and output file!", "pw", 1) return from utilities import get_im d = get_im(args[0]) nx = d.get_xsize() ny = d.get_ysize() if nargs == 3: mask = get_im(args[2]) wn = int(options.wn) if wn == -1: wn = max(nx, ny) else: if ((wn < nx) or (wn < ny)): ERROR("window size cannot be smaller than the image size", "pw", 1) n = EMUtil.get_image_count(args[0]) from utilities import model_blank, model_circle, pad from EMAN2 import periodogram p = model_blank(wn, wn) for i in xrange(n): d = get_im(args[0], i) if nargs == 3: d *= mask st = Util.infomask(d, None, True) d -= st[0] p += periodogram(pad(d, wn, wn, 1, 0.)) p /= n p.write_image(args[1]) elif options.adjpw: if len(args) < 3: ERROR( "filt_by_rops input target output fl aa (the last two are optional parameters of a low-pass filter)", "adjpw", 1) return img_stack = args[0] from math import sqrt from fundamentals import rops_table, fft from utilities import read_text_file, get_im from filter import filt_tanl, filt_table if (args[1][-3:] == 'txt'): rops_dst = read_text_file(args[1]) else: rops_dst = rops_table(get_im(args[1])) out_stack = args[2] if (len(args) > 4): fl = float(args[3]) aa = float(args[4]) else: fl = -1.0 aa = 0.0 nimage = EMUtil.get_image_count(img_stack) for i in xrange(nimage): img = fft(get_im(img_stack, i)) rops_src = rops_table(img) assert len(rops_dst) == len(rops_src) table = [0.0] * len(rops_dst) for j in xrange(len(rops_dst)): table[j] = sqrt(rops_dst[j] / rops_src[j]) if (fl > 0.0): img = filt_tanl(img, fl, aa) img = fft(filt_table(img, table)) img.write_image(out_stack, i) elif options.rotpw != None: if len(args) != 1: ERROR("Only one input permitted", "rotpw", 1) return from utilities import write_text_file, get_im from fundamentals import rops_table from math import log10 t = rops_table(get_im(args[0])) x = range(len(t)) r = [0.0] * len(x) for i in x: r[i] = log10(t[i]) write_text_file([t, r, x], options.rotpw) elif options.transformparams != None: if len(args) != 2: ERROR( "Please provide names of input and output files with orientation parameters", "transformparams", 1) return from utilities import read_text_row, write_text_row transf = [0.0] * 6 spl = options.transformparams.split(',') for i in xrange(len(spl)): transf[i] = float(spl[i]) write_text_row(rotate_shift_params(read_text_row(args[0]), transf), args[1]) elif options.makedb != None: nargs = len(args) if nargs != 1: print "must provide exactly one argument denoting database key under which the input params will be stored" return dbkey = args[0] print "database key under which params will be stored: ", dbkey gbdb = js_open_dict("e2boxercache/gauss_box_DB.json") parmstr = 'dummy:' + options.makedb[0] (processorname, param_dict) = parsemodopt(parmstr) dbdict = {} for pkey in param_dict: if (pkey == 'invert_contrast') or (pkey == 'use_variance'): if param_dict[pkey] == 'True': dbdict[pkey] = True else: dbdict[pkey] = False else: dbdict[pkey] = param_dict[pkey] gbdb[dbkey] = dbdict elif options.generate_projections: nargs = len(args) if nargs != 3: ERROR("Must provide name of input structure(s) from which to generate projections, name of output projection stack, and prefix for output micrographs."\ "sxprocess - generate projections",1) return inpstr = args[0] outstk = args[1] micpref = args[2] parmstr = 'dummy:' + options.generate_projections[0] (processorname, param_dict) = parsemodopt(parmstr) parm_CTF = False parm_format = 'bdb' parm_apix = 2.5 if 'CTF' in param_dict: if param_dict['CTF'] == 'True': parm_CTF = True if 'format' in param_dict: parm_format = param_dict['format'] if 'apix' in param_dict: parm_apix = float(param_dict['apix']) boxsize = 64 if 'boxsize' in param_dict: boxsize = int(param_dict['boxsize']) print "pixel size: ", parm_apix, " format: ", parm_format, " add CTF: ", parm_CTF, " box size: ", boxsize scale_mult = 2500 sigma_add = 1.5 sigma_proj = 30.0 sigma2_proj = 17.5 sigma_gauss = 0.3 sigma_mic = 30.0 sigma2_mic = 17.5 sigma_gauss_mic = 0.3 if 'scale_mult' in param_dict: scale_mult = float(param_dict['scale_mult']) if 'sigma_add' in param_dict: sigma_add = float(param_dict['sigma_add']) if 'sigma_proj' in param_dict: sigma_proj = float(param_dict['sigma_proj']) if 'sigma2_proj' in param_dict: sigma2_proj = float(param_dict['sigma2_proj']) if 'sigma_gauss' in param_dict: sigma_gauss = float(param_dict['sigma_gauss']) if 'sigma_mic' in param_dict: sigma_mic = float(param_dict['sigma_mic']) if 'sigma2_mic' in param_dict: sigma2_mic = float(param_dict['sigma2_mic']) if 'sigma_gauss_mic' in param_dict: sigma_gauss_mic = float(param_dict['sigma_gauss_mic']) from filter import filt_gaussl, filt_ctf from utilities import drop_spider_doc, even_angles, model_gauss, delete_bdb, model_blank, pad, model_gauss_noise, set_params2D, set_params_proj from projection import prep_vol, prgs seed(14567) delta = 29 angles = even_angles(delta, 0.0, 89.9, 0.0, 359.9, "S") nangle = len(angles) modelvol = [] nvlms = EMUtil.get_image_count(inpstr) from utilities import get_im for k in xrange(nvlms): modelvol.append(get_im(inpstr, k)) nx = modelvol[0].get_xsize() if nx != boxsize: ERROR("Requested box dimension does not match dimension of the input model.", \ "sxprocess - generate projections",1) nvol = 10 volfts = [[] for k in xrange(nvlms)] for k in xrange(nvlms): for i in xrange(nvol): sigma = sigma_add + random() # 1.5-2.5 addon = model_gauss(sigma, boxsize, boxsize, boxsize, sigma, sigma, 38, 38, 40) scale = scale_mult * (0.5 + random()) vf, kb = prep_vol(modelvol[k] + scale * addon) volfts[k].append(vf) del vf, modelvol if parm_format == "bdb": stack_data = "bdb:" + outstk delete_bdb(stack_data) else: stack_data = outstk + ".hdf" Cs = 2.0 pixel = parm_apix voltage = 120.0 ampcont = 10.0 ibd = 4096 / 2 - boxsize iprj = 0 width = 240 xstart = 8 + boxsize / 2 ystart = 8 + boxsize / 2 rowlen = 17 from random import randint params = [] for idef in xrange(3, 8): irow = 0 icol = 0 mic = model_blank(4096, 4096) defocus = idef * 0.5 #0.2 if parm_CTF: astampl = defocus * 0.15 astangl = 50.0 ctf = generate_ctf([ defocus, Cs, voltage, pixel, ampcont, 0.0, astampl, astangl ]) for i in xrange(nangle): for k in xrange(12): dphi = 8.0 * (random() - 0.5) dtht = 8.0 * (random() - 0.5) psi = 360.0 * random() phi = angles[i][0] + dphi tht = angles[i][1] + dtht s2x = 4.0 * (random() - 0.5) s2y = 4.0 * (random() - 0.5) params.append([phi, tht, psi, s2x, s2y]) ivol = iprj % nvol #imgsrc = randint(0,nvlms-1) imgsrc = iprj % nvlms proj = prgs(volfts[imgsrc][ivol], kb, [phi, tht, psi, -s2x, -s2y]) x = xstart + irow * width y = ystart + icol * width mic += pad(proj, 4096, 4096, 1, 0.0, x - 2048, y - 2048, 0) proj = proj + model_gauss_noise(sigma_proj, nx, nx) if parm_CTF: proj = filt_ctf(proj, ctf) proj.set_attr_dict({"ctf": ctf, "ctf_applied": 0}) proj = proj + filt_gaussl( model_gauss_noise(sigma2_proj, nx, nx), sigma_gauss) proj.set_attr("origimgsrc", imgsrc) proj.set_attr("test_id", iprj) # flags describing the status of the image (1 = true, 0 = false) set_params2D(proj, [0.0, 0.0, 0.0, 0, 1.0]) set_params_proj(proj, [phi, tht, psi, s2x, s2y]) proj.write_image(stack_data, iprj) icol += 1 if icol == rowlen: icol = 0 irow += 1 iprj += 1 mic += model_gauss_noise(sigma_mic, 4096, 4096) if parm_CTF: #apply CTF mic = filt_ctf(mic, ctf) mic += filt_gaussl(model_gauss_noise(sigma2_mic, 4096, 4096), sigma_gauss_mic) mic.write_image(micpref + "%1d.hdf" % (idef - 3), 0) drop_spider_doc("params.txt", params) elif options.importctf != None: print ' IMPORTCTF ' from utilities import read_text_row, write_text_row from random import randint import subprocess grpfile = 'groupid%04d' % randint(1000, 9999) ctfpfile = 'ctfpfile%04d' % randint(1000, 9999) cterr = [options.defocuserror / 100.0, options.astigmatismerror] ctfs = read_text_row(options.importctf) for kk in xrange(len(ctfs)): root, name = os.path.split(ctfs[kk][-1]) ctfs[kk][-1] = name[:-4] if (options.input[:4] != 'bdb:'): ERROR('Sorry, only bdb files implemented', 'importctf', 1) d = options.input[4:] #try: str = d.index('*') #except: str = -1 from string import split import glob uu = os.path.split(d) uu = os.path.join(uu[0], 'EMAN2DB', uu[1] + '.bdb') flist = glob.glob(uu) for i in xrange(len(flist)): root, name = os.path.split(flist[i]) root = root[:-7] name = name[:-4] fil = 'bdb:' + os.path.join(root, name) sourcemic = EMUtil.get_all_attributes(fil, 'ptcl_source_image') nn = len(sourcemic) gctfp = [] groupid = [] for kk in xrange(nn): junk, name2 = os.path.split(sourcemic[kk]) name2 = name2[:-4] ctfp = [-1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0] for ll in xrange(len(ctfs)): if (name2 == ctfs[ll][-1]): # found correct if (ctfs[ll][8] / ctfs[ll][0] <= cterr[0]): # acceptable defocus error ctfp = ctfs[ll][:8] if (ctfs[ll][10] > cterr[1]): # error of astigmatism exceed the threshold, set astigmatism to zero. ctfp[6] = 0.0 ctfp[7] = 0.0 gctfp.append(ctfp) groupid.append(kk) break if (len(groupid) > 0): write_text_row(groupid, grpfile) write_text_row(gctfp, ctfpfile) cmd = "{} {} {} {}".format( 'e2bdb.py', fil, '--makevstack=bdb:' + root + 'G' + name, '--list=' + grpfile) #print cmd subprocess.call(cmd, shell=True) cmd = "{} {} {} {}".format('sxheader.py', 'bdb:' + root + 'G' + name, '--params=ctf', '--import=' + ctfpfile) #print cmd subprocess.call(cmd, shell=True) else: print ' >>> Group ', name, ' skipped.' cmd = "{} {} {}".format("rm -f", grpfile, ctfpfile) subprocess.call(cmd, shell=True) elif options.scale > 0.0: from utilities import read_text_row, write_text_row scale = options.scale nargs = len(args) if nargs != 2: print "Please provide names of input and output file!" return p = read_text_row(args[0]) for i in xrange(len(p)): p[i][3] /= scale p[i][4] /= scale write_text_row(p, args[1]) elif options.adaptive_mask: from utilities import get_im from morphology import adaptive_mask nsigma = options.nsigma ndilation = options.ndilation kernel_size = options.kernel_size gauss_standard_dev = options.gauss_standard_dev nargs = len(args) if nargs > 2: print "Too many inputs are given, try again!" return else: inputvol = get_im(args[0]) input_path, input_file_name = os.path.split(args[0]) input_file_name_root, ext = os.path.splitext(input_file_name) if nargs == 2: mask_file_name = args[1] else: mask_file_name = "adaptive_mask_for" + input_file_name_root + ".hdf" # Only hdf file is output. mask3d = adaptive_mask(inputvol, nsigma, ndilation, kernel_size, gauss_standard_dev) mask3d.write_image(mask_file_name) else: ERROR("Please provide option name", "sxprocess.py", 1)
def main(): import sys import os import math import random import pyemtbx.options import time from random import random, seed, randint from optparse import OptionParser progname = os.path.basename(sys.argv[0]) usage = progname + """ [options] <inputfile> <outputfile> Generic 2-D image processing programs. Functionality: 1. Phase flip a stack of images and write output to new file: sxprocess.py input_stack.hdf output_stack.hdf --phase_flip 2. Resample (decimate or interpolate up) images (2D or 3D) in a stack to change the pixel size. The window size will change accordingly. sxprocess input.hdf output.hdf --changesize --ratio=0.5 3. Compute average power spectrum of a stack of 2D images with optional padding (option wn) with zeroes or a 3-D volume. sxprocess.py input_stack.hdf powerspectrum.hdf --pw [--wn=1024] 4. Generate a stack of projections bdb:data and micrographs with prefix mic (i.e., mic0.hdf, mic1.hdf etc) from structure input_structure.hdf, with CTF applied to both projections and micrographs: sxprocess.py input_structure.hdf data mic --generate_projections format="bdb":apix=5.2:CTF=True:boxsize=64 5. Retrieve original image numbers in the selected ISAC group (here group 12 from generation 3): sxprocess.py bdb:test3 class_averages_generation_3.hdf list3_12.txt --isacgroup=12 --params=originalid 6. Retrieve original image numbers of images listed in ISAC output stack of averages: sxprocess.py select1.hdf ohk.txt 7. Adjust rotationally averaged power spectrum of an image to that of a reference image or a reference 1D power spectrum stored in an ASCII file. Optionally use a tangent low-pass filter. Also works for a stack of images, in which case the output is also a stack. sxprocess.py vol.hdf ref.hdf avol.hdf < 0.25 0.2> --adjpw sxprocess.py vol.hdf pw.txt avol.hdf < 0.25 0.2> --adjpw 8. Generate a 1D rotationally averaged power spectrum of an image. sxprocess.py vol.hdf --rotwp=rotpw.txt # Output will contain three columns: (1) rotationally averaged power spectrum (2) logarithm of the rotationally averaged power spectrum (3) integer line number (from zero to approximately to half the image size) 9. Apply 3D transformation (rotation and/or shift) to a set of orientation parameters associated with projection data. sxprocess.py --transfromparams=phi,theta,psi,tx,ty,tz input.txt output.txt The output file is then imported and 3D transformed volume computed: sxheader.py bdb:p --params=xform.projection --import=output.txt mpirun -np 2 sxrecons3d_n.py bdb:p tvol.hdf --MPI The reconstructed volume is in the position of the volume computed using the input.txt parameters and then transformed with rot_shift3D(vol, phi,theta,psi,tx,ty,tz) 10. Import ctf parameters from the output of sxcter into windowed particle headers. There are three possible input files formats: (1) all particles are in one stack, (2 aor 3) particles are in stacks, each stack corresponds to a single micrograph. In each case the particles should contain a name of the micrograph of origin stores using attribute name 'ptcl_source_image'. Normally this is done by e2boxer.py during windowing. Particles whose defocus or astigmatism error exceed set thresholds will be skipped, otherwise, virtual stacks with the original way preceded by G will be created. sxprocess.py --input=bdb:data --importctf=outdir/partres --defocuserror=10.0 --astigmatismerror=5.0 # Output will be a vritual stack bdb:Gdata sxprocess.py --input="bdb:directory/stacks*" --importctf=outdir/partres --defocuserror=10.0 --astigmatismerror=5.0 To concatenate output files: cd directory e2bdb.py . --makevstack=bdb:allparticles --filt=G IMPORTANT: Please do not move (or remove!) any input/intermediate EMAN2DB files as the information is linked between them. 11. Scale 3D shifts. The shifts in the input five columns text file with 3D orientation parameters will be DIVIDED by the scale factor sxprocess.py orientationparams.txt scaledparams.txt scale=0.5 12. Generate 3D mask from a given 3-D volume automatically or using threshold provided by user. 13. Postprocess 3-D or 2-D images: for 3-D volumes: calculate FSC with provided mask; weight summed volume with FSC; estimate B-factor from FSC weighted summed two volumes; apply negative B-factor to the weighted volume. for 2-D images: calculate B-factor and apply negative B-factor to 2-D images. 14. Winow stack file -reduce size of images without changing the pixel size. """ parser = OptionParser(usage,version=SPARXVERSION) parser.add_option("--order", action="store_true", help="Two arguments are required: name of input stack and desired name of output stack. The output stack is the input stack sorted by similarity in terms of cross-correlation coefficent.", default=False) parser.add_option("--order_lookup", action="store_true", help="Test/Debug.", default=False) parser.add_option("--order_metropolis", action="store_true", help="Test/Debug.", default=False) parser.add_option("--order_pca", action="store_true", help="Test/Debug.", default=False) parser.add_option("--initial", type="int", default=-1, help="Specifies which image will be used as an initial seed to form the chain. (default = 0, means the first image)") parser.add_option("--circular", action="store_true", help="Select circular ordering (fisr image has to be similar to the last", default=False) parser.add_option("--radius", type="int", default=-1, help="Radius of a circular mask for similarity based ordering") parser.add_option("--changesize", action="store_true", help="resample (decimate or interpolate up) images (2D or 3D) in a stack to change the pixel size.", default=False) parser.add_option("--ratio", type="float", default=1.0, help="The ratio of new to old image size (if <1 the pixel size will increase and image size decrease, if>1, the other way round") parser.add_option("--pw", action="store_true", help="compute average power spectrum of a stack of 2-D images with optional padding (option wn) with zeroes", default=False) parser.add_option("--wn", type="int", default=-1, help="Size of window to use (should be larger/equal than particle box size, default padding to max(nx,ny))") parser.add_option("--phase_flip", action="store_true", help="Phase flip the input stack", default=False) parser.add_option("--makedb", metavar="param1=value1:param2=value2", type="string", action="append", help="One argument is required: name of key with which the database will be created. Fill in database with parameters specified as follows: --makedb param1=value1:param2=value2, e.g. 'gauss_width'=1.0:'pixel_input'=5.2:'pixel_output'=5.2:'thr_low'=1.0") parser.add_option("--generate_projections", metavar="param1=value1:param2=value2", type="string", action="append", help="Three arguments are required: name of input structure from which to generate projections, desired name of output projection stack, and desired prefix for micrographs (e.g. if prefix is 'mic', then micrographs mic0.hdf, mic1.hdf etc will be generated). Optional arguments specifying format, apix, box size and whether to add CTF effects can be entered as follows after --generate_projections: format='bdb':apix=5.2:CTF=True:boxsize=100, or format='hdf', etc., where format is bdb or hdf, apix (pixel size) is a float, CTF is True or False, and boxsize denotes the dimension of the box (assumed to be a square). If an optional parameter is not specified, it will default as follows: format='bdb', apix=2.5, CTF=False, boxsize=64.") parser.add_option("--isacgroup", type="int", help="Retrieve original image numbers in the selected ISAC group. See ISAC documentation for details.", default=-1) parser.add_option("--isacselect", action="store_true", help="Retrieve original image numbers of images listed in ISAC output stack of averages. See ISAC documentation for details.", default=False) parser.add_option("--params", type="string", default=None, help="Name of header of parameter, which one depends on specific option") parser.add_option("--adjpw", action="store_true", help="Adjust rotationally averaged power spectrum of an image", default=False) parser.add_option("--rotpw", type="string", default=None, help="Name of the text file to contain rotationally averaged power spectrum of the input image.") parser.add_option("--transformparams", type="string", default=None, help="Transform 3D projection orientation parameters using six 3D parameters (phi, theta,psi,sx,sy,sz). Input: --transformparams=45.,66.,12.,-2,3,-5.5 desired six transformation of the reconstructed structure. Output: file with modified orientation parameters.") # import ctf estimates done using cter parser.add_option("--input", type="string", default= None, help="Input particles.") parser.add_option("--importctf", type="string", default= None, help="Name of the file containing CTF parameters produced by sxcter.") parser.add_option("--defocuserror", type="float", default=1000000.0, help="Exclude micrographs whose relative defocus error as estimated by sxcter is larger than defocuserror percent. The error is computed as (std dev defocus)/defocus*100%") parser.add_option("--astigmatismerror", type="float", default=360.0, help="Set to zero astigmatism for micrographs whose astigmatism angular error as estimated by sxcter is larger than astigmatismerror degrees.") # import ctf estimates done using cter parser.add_option("--scale", type="float", default=-1.0, help="Divide shifts in the input 3D orientation parameters text file by the scale factor.") # generate adaptive mask from an given 3-D volume parser.add_option("--adaptive_mask", action="store_true", help="create adavptive 3-D mask from a given volume", default=False) parser.add_option("--nsigma", type="float", default= 1., help="number of times of sigma of the input volume to obtain the the large density cluster") parser.add_option("--ndilation", type="int", default= 3, help="number of times of dilation applied to the largest cluster of density") parser.add_option("--kernel_size", type="int", default= 11, help="convolution kernel for smoothing the edge of the mask") parser.add_option("--gauss_standard_dev", type="int", default= 9, help="stanadard deviation value to generate Gaussian edge") parser.add_option("--threshold", type="float", default= 9999., help="threshold provided by user to binarize input volume") parser.add_option("--ne", type="int", default= 0, help="number of times to erode the binarized input image") parser.add_option("--nd", type="int", default= 0, help="number of times to dilate the binarized input image") parser.add_option("--postprocess", action="store_true", help="postprocess unfiltered odd, even 3-D volumes",default=False) parser.add_option("--fsc_weighted", action="store_true", help="postprocess unfiltered odd, even 3-D volumes") parser.add_option("--low_pass_filter", action="store_true", default=False, help="postprocess unfiltered odd, even 3-D volumes") parser.add_option("--ff", type="float", default=.25, help="low pass filter stop band frequency in absolute unit") parser.add_option("--aa", type="float", default=.1, help="low pass filter falloff" ) parser.add_option("--mask", type="string", help="input mask file", default=None) parser.add_option("--output", type="string", help="output file name", default="postprocessed.hdf") parser.add_option("--pixel_size", type="float", help="pixel size of the data", default=1.0) parser.add_option("--B_start", type="float", help="starting frequency in Angstrom for B-factor estimation", default=10.) parser.add_option("--FSC_cutoff", type="float", help="stop frequency in Angstrom for B-factor estimation", default=0.143) parser.add_option("--2d", action="store_true", help="postprocess isac 2-D averaged images",default=False) parser.add_option("--window_stack", action="store_true", help="window stack images using a smaller window size", default=False) parser.add_option("--box", type="int", default= 0, help="the new window size ") (options, args) = parser.parse_args() global_def.BATCH = True if options.phase_flip: nargs = len(args) if nargs != 2: print "must provide name of input and output file!" return from EMAN2 import Processor instack = args[0] outstack = args[1] nima = EMUtil.get_image_count(instack) from filter import filt_ctf for i in xrange(nima): img = EMData() img.read_image(instack, i) try: ctf = img.get_attr('ctf') except: print "no ctf information in input stack! Exiting..." return dopad = True sign = 1 binary = 1 # phase flip assert img.get_ysize() > 1 dict = ctf.to_dict() dz = dict["defocus"] cs = dict["cs"] voltage = dict["voltage"] pixel_size = dict["apix"] b_factor = dict["bfactor"] ampcont = dict["ampcont"] dza = dict["dfdiff"] azz = dict["dfang"] if dopad and not img.is_complex(): ip = 1 else: ip = 0 params = {"filter_type": Processor.fourier_filter_types.CTF_, "defocus" : dz, "Cs": cs, "voltage": voltage, "Pixel_size": pixel_size, "B_factor": b_factor, "amp_contrast": ampcont, "dopad": ip, "binary": binary, "sign": sign, "dza": dza, "azz":azz} tmp = Processor.EMFourierFilter(img, params) tmp.set_attr_dict({"ctf": ctf}) tmp.write_image(outstack, i) elif options.changesize: nargs = len(args) if nargs != 2: ERROR("must provide name of input and output file!", "change size", 1) return from utilities import get_im instack = args[0] outstack = args[1] sub_rate = float(options.ratio) nima = EMUtil.get_image_count(instack) from fundamentals import resample for i in xrange(nima): resample(get_im(instack, i), sub_rate).write_image(outstack, i) elif options.isacgroup>-1: nargs = len(args) if nargs != 3: ERROR("Three files needed on input!", "isacgroup", 1) return from utilities import get_im instack = args[0] m=get_im(args[1],int(options.isacgroup)).get_attr("members") l = [] for k in m: l.append(int(get_im(args[0],k).get_attr(options.params))) from utilities import write_text_file write_text_file(l, args[2]) elif options.isacselect: nargs = len(args) if nargs != 2: ERROR("Two files needed on input!", "isacgroup", 1) return from utilities import get_im nima = EMUtil.get_image_count(args[0]) m = [] for k in xrange(nima): m += get_im(args[0],k).get_attr("members") m.sort() from utilities import write_text_file write_text_file(m, args[1]) elif options.pw: nargs = len(args) if nargs < 2: ERROR("must provide name of input and output file!", "pw", 1) return from utilities import get_im, write_text_file from fundamentals import rops_table d = get_im(args[0]) ndim = d.get_ndim() if ndim ==3: pw = rops_table(d) write_text_file(pw, args[1]) else: nx = d.get_xsize() ny = d.get_ysize() if nargs ==3: mask = get_im(args[2]) wn = int(options.wn) if wn == -1: wn = max(nx, ny) else: if( (wn<nx) or (wn<ny) ): ERROR("window size cannot be smaller than the image size","pw",1) n = EMUtil.get_image_count(args[0]) from utilities import model_blank, model_circle, pad from EMAN2 import periodogram p = model_blank(wn,wn) for i in xrange(n): d = get_im(args[0], i) if nargs==3: d *=mask st = Util.infomask(d, None, True) d -= st[0] p += periodogram(pad(d, wn, wn, 1, 0.)) p /= n p.write_image(args[1]) elif options.adjpw: if len(args) < 3: ERROR("filt_by_rops input target output fl aa (the last two are optional parameters of a low-pass filter)","adjpw",1) return img_stack = args[0] from math import sqrt from fundamentals import rops_table, fft from utilities import read_text_file, get_im from filter import filt_tanl, filt_table if( args[1][-3:] == 'txt'): rops_dst = read_text_file( args[1] ) else: rops_dst = rops_table(get_im( args[1] )) out_stack = args[2] if(len(args) >4): fl = float(args[3]) aa = float(args[4]) else: fl = -1.0 aa = 0.0 nimage = EMUtil.get_image_count( img_stack ) for i in xrange(nimage): img = fft(get_im(img_stack, i) ) rops_src = rops_table(img) assert len(rops_dst) == len(rops_src) table = [0.0]*len(rops_dst) for j in xrange( len(rops_dst) ): table[j] = sqrt( rops_dst[j]/rops_src[j] ) if( fl > 0.0): img = filt_tanl(img, fl, aa) img = fft(filt_table(img, table)) img.write_image(out_stack, i) elif options.rotpw != None: if len(args) != 1: ERROR("Only one input permitted","rotpw",1) return from utilities import write_text_file, get_im from fundamentals import rops_table from math import log10 t = rops_table(get_im(args[0])) x = range(len(t)) r = [0.0]*len(x) for i in x: r[i] = log10(t[i]) write_text_file([t,r,x],options.rotpw) elif options.transformparams != None: if len(args) != 2: ERROR("Please provide names of input and output files with orientation parameters","transformparams",1) return from utilities import read_text_row, write_text_row transf = [0.0]*6 spl=options.transformparams.split(',') for i in xrange(len(spl)): transf[i] = float(spl[i]) write_text_row( rotate_shift_params(read_text_row(args[0]), transf) , args[1]) elif options.makedb != None: nargs = len(args) if nargs != 1: print "must provide exactly one argument denoting database key under which the input params will be stored" return dbkey = args[0] print "database key under which params will be stored: ", dbkey gbdb = js_open_dict("e2boxercache/gauss_box_DB.json") parmstr = 'dummy:'+options.makedb[0] (processorname, param_dict) = parsemodopt(parmstr) dbdict = {} for pkey in param_dict: if (pkey == 'invert_contrast') or (pkey == 'use_variance'): if param_dict[pkey] == 'True': dbdict[pkey] = True else: dbdict[pkey] = False else: dbdict[pkey] = param_dict[pkey] gbdb[dbkey] = dbdict elif options.generate_projections: nargs = len(args) if nargs != 3: ERROR("Must provide name of input structure(s) from which to generate projections, name of output projection stack, and prefix for output micrographs."\ "sxprocess - generate projections",1) return inpstr = args[0] outstk = args[1] micpref = args[2] parmstr = 'dummy:'+options.generate_projections[0] (processorname, param_dict) = parsemodopt(parmstr) parm_CTF = False parm_format = 'bdb' parm_apix = 2.5 if 'CTF' in param_dict: if param_dict['CTF'] == 'True': parm_CTF = True if 'format' in param_dict: parm_format = param_dict['format'] if 'apix' in param_dict: parm_apix = float(param_dict['apix']) boxsize = 64 if 'boxsize' in param_dict: boxsize = int(param_dict['boxsize']) print "pixel size: ", parm_apix, " format: ", parm_format, " add CTF: ", parm_CTF, " box size: ", boxsize scale_mult = 2500 sigma_add = 1.5 sigma_proj = 30.0 sigma2_proj = 17.5 sigma_gauss = 0.3 sigma_mic = 30.0 sigma2_mic = 17.5 sigma_gauss_mic = 0.3 if 'scale_mult' in param_dict: scale_mult = float(param_dict['scale_mult']) if 'sigma_add' in param_dict: sigma_add = float(param_dict['sigma_add']) if 'sigma_proj' in param_dict: sigma_proj = float(param_dict['sigma_proj']) if 'sigma2_proj' in param_dict: sigma2_proj = float(param_dict['sigma2_proj']) if 'sigma_gauss' in param_dict: sigma_gauss = float(param_dict['sigma_gauss']) if 'sigma_mic' in param_dict: sigma_mic = float(param_dict['sigma_mic']) if 'sigma2_mic' in param_dict: sigma2_mic = float(param_dict['sigma2_mic']) if 'sigma_gauss_mic' in param_dict: sigma_gauss_mic = float(param_dict['sigma_gauss_mic']) from filter import filt_gaussl, filt_ctf from utilities import drop_spider_doc, even_angles, model_gauss, delete_bdb, model_blank,pad,model_gauss_noise,set_params2D, set_params_proj from projection import prep_vol,prgs seed(14567) delta = 29 angles = even_angles(delta, 0.0, 89.9, 0.0, 359.9, "S") nangle = len(angles) modelvol = [] nvlms = EMUtil.get_image_count(inpstr) from utilities import get_im for k in xrange(nvlms): modelvol.append(get_im(inpstr,k)) nx = modelvol[0].get_xsize() if nx != boxsize: ERROR("Requested box dimension does not match dimension of the input model.", \ "sxprocess - generate projections",1) nvol = 10 volfts = [[] for k in xrange(nvlms)] for k in xrange(nvlms): for i in xrange(nvol): sigma = sigma_add + random() # 1.5-2.5 addon = model_gauss(sigma, boxsize, boxsize, boxsize, sigma, sigma, 38, 38, 40 ) scale = scale_mult * (0.5+random()) vf, kb = prep_vol(modelvol[k] + scale*addon) volfts[k].append(vf) del vf, modelvol if parm_format == "bdb": stack_data = "bdb:"+outstk delete_bdb(stack_data) else: stack_data = outstk + ".hdf" Cs = 2.0 pixel = parm_apix voltage = 120.0 ampcont = 10.0 ibd = 4096/2-boxsize iprj = 0 width = 240 xstart = 8 + boxsize/2 ystart = 8 + boxsize/2 rowlen = 17 from random import randint params = [] for idef in xrange(3, 8): irow = 0 icol = 0 mic = model_blank(4096, 4096) defocus = idef * 0.5#0.2 if parm_CTF: astampl=defocus*0.15 astangl=50.0 ctf = generate_ctf([defocus, Cs, voltage, pixel, ampcont, 0.0, astampl, astangl]) for i in xrange(nangle): for k in xrange(12): dphi = 8.0*(random()-0.5) dtht = 8.0*(random()-0.5) psi = 360.0*random() phi = angles[i][0]+dphi tht = angles[i][1]+dtht s2x = 4.0*(random()-0.5) s2y = 4.0*(random()-0.5) params.append([phi, tht, psi, s2x, s2y]) ivol = iprj % nvol #imgsrc = randint(0,nvlms-1) imgsrc = iprj % nvlms proj = prgs(volfts[imgsrc][ivol], kb, [phi, tht, psi, -s2x, -s2y]) x = xstart + irow * width y = ystart + icol * width mic += pad(proj, 4096, 4096, 1, 0.0, x-2048, y-2048, 0) proj = proj + model_gauss_noise( sigma_proj, nx, nx ) if parm_CTF: proj = filt_ctf(proj, ctf) proj.set_attr_dict({"ctf":ctf, "ctf_applied":0}) proj = proj + filt_gaussl(model_gauss_noise(sigma2_proj, nx, nx), sigma_gauss) proj.set_attr("origimgsrc",imgsrc) proj.set_attr("test_id", iprj) # flags describing the status of the image (1 = true, 0 = false) set_params2D(proj, [0.0, 0.0, 0.0, 0, 1.0]) set_params_proj(proj, [phi, tht, psi, s2x, s2y]) proj.write_image(stack_data, iprj) icol += 1 if icol == rowlen: icol = 0 irow += 1 iprj += 1 mic += model_gauss_noise(sigma_mic,4096,4096) if parm_CTF: #apply CTF mic = filt_ctf(mic, ctf) mic += filt_gaussl(model_gauss_noise(sigma2_mic, 4096, 4096), sigma_gauss_mic) mic.write_image(micpref + "%1d.hdf" % (idef-3), 0) drop_spider_doc("params.txt", params) elif options.importctf != None: print ' IMPORTCTF ' from utilities import read_text_row,write_text_row from random import randint import subprocess grpfile = 'groupid%04d'%randint(1000,9999) ctfpfile = 'ctfpfile%04d'%randint(1000,9999) cterr = [options.defocuserror/100.0, options.astigmatismerror] ctfs = read_text_row(options.importctf) for kk in xrange(len(ctfs)): root,name = os.path.split(ctfs[kk][-1]) ctfs[kk][-1] = name[:-4] if(options.input[:4] != 'bdb:'): ERROR('Sorry, only bdb files implemented','importctf',1) d = options.input[4:] #try: str = d.index('*') #except: str = -1 from string import split import glob uu = os.path.split(d) uu = os.path.join(uu[0],'EMAN2DB',uu[1]+'.bdb') flist = glob.glob(uu) for i in xrange(len(flist)): root,name = os.path.split(flist[i]) root = root[:-7] name = name[:-4] fil = 'bdb:'+os.path.join(root,name) sourcemic = EMUtil.get_all_attributes(fil,'ptcl_source_image') nn = len(sourcemic) gctfp = [] groupid = [] for kk in xrange(nn): junk,name2 = os.path.split(sourcemic[kk]) name2 = name2[:-4] ctfp = [-1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0] for ll in xrange(len(ctfs)): if(name2 == ctfs[ll][-1]): # found correct if(ctfs[ll][8]/ctfs[ll][0] <= cterr[0]): # acceptable defocus error ctfp = ctfs[ll][:8] if(ctfs[ll][10] > cterr[1] ): # error of astigmatism exceed the threshold, set astigmatism to zero. ctfp[6] = 0.0 ctfp[7] = 0.0 gctfp.append(ctfp) groupid.append(kk) break if(len(groupid) > 0): write_text_row(groupid, grpfile) write_text_row(gctfp, ctfpfile) cmd = "{} {} {} {}".format('e2bdb.py',fil,'--makevstack=bdb:'+root+'G'+name,'--list='+grpfile) #print cmd subprocess.call(cmd, shell=True) cmd = "{} {} {} {}".format('sxheader.py','bdb:'+root+'G'+name,'--params=ctf','--import='+ctfpfile) #print cmd subprocess.call(cmd, shell=True) else: print ' >>> Group ',name,' skipped.' cmd = "{} {} {}".format("rm -f",grpfile,ctfpfile) subprocess.call(cmd, shell=True) elif options.scale > 0.0: from utilities import read_text_row,write_text_row scale = options.scale nargs = len(args) if nargs != 2: print "Please provide names of input and output file!" return p = read_text_row(args[0]) for i in xrange(len(p)): p[i][3] /= scale p[i][4] /= scale write_text_row(p, args[1]) elif options.adaptive_mask: from utilities import get_im from morphology import adaptive_mask, binarize, erosion, dilation nsigma = options.nsigma ndilation = options.ndilation kernel_size = options.kernel_size gauss_standard_dev = options.gauss_standard_dev nargs = len(args) if nargs ==0: print " Create 3D mask from a given volume, either automatically or from the user provided threshold." elif nargs > 2: print "Too many inputs are given, try again!" return else: inputvol = get_im(args[0]) input_path, input_file_name = os.path.split(args[0]) input_file_name_root,ext=os.path.splitext(input_file_name) if nargs == 2: mask_file_name = args[1] else: mask_file_name = "adaptive_mask_for_"+input_file_name_root+".hdf" # Only hdf file is output. if options.threshold !=9999.: mask3d = binarize(inputvol, options.threshold) for i in xrange(options.ne): mask3d = erosion(mask3d) for i in xrange(options.nd): mask3d = dilation(mask3d) else: mask3d = adaptive_mask(inputvol, nsigma, ndilation, kernel_size, gauss_standard_dev) mask3d.write_image(mask_file_name) elif options.postprocess: from utilities import get_im from fundamentals import rot_avg_table from morphology import compute_bfactor,power from statistics import fsc from filter import filt_table, filt_gaussinv from EMAN2 import periodogram e1 = get_im(args[0],0) if e1.get_zsize()==1: nimage = EMUtil.get_image_count(args[0]) if options.mask !=None: m = get_im(options.mask) else: m = None for i in xrange(nimage): e1 = get_im(args[0],i) if m: e1 *=m guinerline = rot_avg_table(power(periodogram(e1),.5)) freq_max = 1/(2.*pixel_size) freq_min = 1./options.B_start b,junk=compute_bfactor(guinerline, freq_min, freq_max, pixel_size) tmp = b/pixel_size**2 sigma_of_inverse=sqrt(2./tmp) e1 = filt_gaussinv(e1,sigma_of_inverse) if options.low_pass_filter: from filter import filt_tanl e1 =filt_tanl(e1,options.ff, options.aa) e1.write_image(options.output) else: nargs = len(args) e1 = get_im(args[0]) if nargs >1: e2 = get_im(args[1]) if options.mask !=None: m = get_im(options.mask) else: m =None pixel_size = options.pixel_size from math import sqrt if m !=None: e1 *=m if nargs >1 :e2 *=m if options.fsc_weighted: frc = fsc(e1,e2,1) ## FSC is done on masked two images #### FSC weighting sqrt((2.*fsc)/(1+fsc)); fil = len(frc[1])*[None] for i in xrange(len(fil)): if frc[1][i]>=options.FSC_cutoff: tmp = frc[1][i] else: tmp = 0.0 fil[i] = sqrt(2.*tmp/(1.+tmp)) if nargs>1: e1 +=e2 if options.fsc_weighted: e1=filt_table(e1,fil) guinerline = rot_avg_table(power(periodogram(e1),.5)) freq_max = 1/(2.*pixel_size) freq_min = 1./options.B_start b,junk = compute_bfactor(guinerline, freq_min, freq_max, pixel_size) tmp = b/pixel_size**2 sigma_of_inverse=sqrt(2./tmp) e1 = filt_gaussinv(e1,sigma_of_inverse) if options.low_pass_filter: from filter import filt_tanl e1 =filt_tanl(e1,options.ff, options.aa) e1.write_image(options.output) elif options.window_stack: nargs = len(args) if nargs ==0: print " Reduce image size of a stack" return else: output_stack_name = None inputstack = args[0] if nargs ==2:output_stack_name = args[1] input_path,input_file_name=os.path.split(inputstack) input_file_name_root,ext=os.path.splitext(input_file_name) if input_file_name_root[0:3]=="bdb":stack_is_bdb= True else: stack_is_bdb= False if output_stack_name is None: if stack_is_bdb: output_stack_name ="bdb:reduced_"+input_file_name_root[4:] else:output_stack_name = "reduced_"+input_file_name_root+".hdf" # Only hdf file is output. nimage = EMUtil.get_image_count(inputstack) from fundamentals import window2d for i in xrange(nimage): image = EMData() image.read_image(inputstack,i) w = window2d(image,options.box,options.box) w.write_image(output_stack_name,i) else: ERROR("Please provide option name","sxprocess.py",1)
def main(): from utilities import write_text_row, drop_image, model_gauss_noise, get_im, set_params_proj, wrap_mpi_bcast, model_circle import user_functions from applications import MPI_start_end from optparse import OptionParser from global_def import SPARXVERSION from EMAN2 import EMData from multi_shc import multi_shc, do_volume from logger import Logger, BaseLogger_Files import sys import os import time import socket progname = os.path.basename(sys.argv[0]) usage = progname + " stack [output_directory] initial_volume --ir=inner_radius --ou=outer_radius --rs=ring_step --xr=x_range --yr=y_range --ts=translational_search_step --delta=angular_step --an=angular_neighborhood --center=center_type --fl --aa --ref_a=S --sym=c1" parser = OptionParser(usage,version=SPARXVERSION) parser.add_option("--ir", type= "int", default= 1, help="inner radius for rotational correlation > 0 (set to 1)") parser.add_option("--ou", type= "int", default= -1, help="outer radius for rotational correlation < int(nx/2)-1 (set to the radius of the particle)") parser.add_option("--rs", type= "int", default= 1, help="step between rings in rotational correlation >0 (set to 1)" ) parser.add_option("--xr", type="string", default= "-1", help="range for translation search in x direction, search is +/xr (default 0)") parser.add_option("--yr", type="string", default= "-1", help="range for translation search in y direction, search is +/yr (default = same as xr)") parser.add_option("--ts", type="string", default= "1", help="step size of the translation search in both directions, search is -xr, -xr+ts, 0, xr-ts, xr, can be fractional") parser.add_option("--delta", type="string", default= "-1", help="angular step of reference projections during initialization step (default automatically selected based on radius of the structure.)") #parser.add_option("--an", type="string", default= "-1", help="angular neighborhood for local searches (phi and theta)") parser.add_option("--center", type="float", default= -1, help="-1: average shift method; 0: no centering; 1: center of gravity (default=-1)") parser.add_option("--maxit", type="int", default= 400, help="maximum number of iterations performed for the GA part (set to 400) ") parser.add_option("--outlier_percentile",type="float", default= 95, help="percentile above which outliers are removed every iteration") parser.add_option("--iteration_start",type="int", default= 0, help="starting iteration for rviper, 0 means go to the most recent one (default).") parser.add_option("--CTF", action="store_true", default=False, help="Use CTF (Default no CTF correction)") parser.add_option("--snr", type="float", default= 1.0, help="Signal-to-Noise Ratio of the data (default 1.0)") parser.add_option("--ref_a", type="string", default= "S", help="method for generating the quasi-uniformly distributed projection directions (default S)") parser.add_option("--sym", type="string", default= "c1", help="symmetry of the refined structure") parser.add_option("--npad", type="int", default= 2, help="padding size for 3D reconstruction (default=2)") parser.add_option("--startangles", action="store_true", default=False, help="Use orientation parameters in the input file header to jumpstart the procedure") #options introduced for the do_volume function parser.add_option("--fl", type="float", default=0.12, help="cut-off frequency of hyperbolic tangent low-pass Fourier filte (default 0.12)") parser.add_option("--aa", type="float", default=0.1, help="fall-off of hyperbolic tangent low-pass Fourier filter (default 0.1)") parser.add_option("--pwreference", type="string", default="", help="text file with a reference power spectrum (default no power spectrum adjustment)") parser.add_option("--mask3D", type="string", default=None, help="3D mask file (default a sphere WHAT RADIUS??)") (options, args) = parser.parse_args(sys.argv[1:]) #print( " args ",args) if( len(args) == 3): volinit = args[2] masterdir = args[1] elif(len(args) == 2): volinit = args[1] masterdir = "" else: print( "usage: " + usage) print( "Please run '" + progname + " -h' for detailed options") return 1 orgstack = args[0] #print( orgstack,masterdir,volinit ) # INPUT PARAMETERS radi = options.ou global_def.BATCH = True ali3d_options.ir = options.ir ali3d_options.rs = options.rs ali3d_options.ou = options.ou ali3d_options.xr = options.xr ali3d_options.yr = options.yr ali3d_options.ts = options.ts ali3d_options.an = "-1" ali3d_options.sym = options.sym ali3d_options.delta = options.delta ali3d_options.npad = options.npad ali3d_options.center = options.center ali3d_options.CTF = options.CTF ali3d_options.ref_a = options.ref_a ali3d_options.snr = options.snr ali3d_options.mask3D = options.mask3D ali3d_options.pwreference = options.pwreference ali3d_options.fl = 0.4 ali3d_options.aa = 0.1 if( ali3d_options.xr == "-1" ): ali3d_options.xr = "2" """ print( options) print( 'ali3d_options', ali3d_options.ir ,\ ali3d_options.rs ,\ ali3d_options.ou ,\ ali3d_options.xr ,\ ali3d_options.yr ,\ ali3d_options.ts ,\ ali3d_options.an ,\ ali3d_options.sym ,\ ali3d_options.delta ,\ ali3d_options.npad ,\ ali3d_options.center ,\ ali3d_options.CTF ,\ ali3d_options.ref_a ,\ ali3d_options.snr ,\ ali3d_options.mask3D ,\ ali3d_options.fl ,\ ali3d_options.aa \ ) #exit() """ mpi_init(0, []) nproc = mpi_comm_size(MPI_COMM_WORLD) myid = mpi_comm_rank(MPI_COMM_WORLD) main_node = 0 #mpi_finalize() #exit() nxinit = -1 #int(280*0.3*2) nsoft = 0 mempernode = 4.0e9 # PARAMETERS OF THE PROCEDURE # threshold error thresherr = 0 fq = 0.11 # low-freq limit to which fuse ref volumes. Should it be estimated somehow? # Get the pixel size, if none set to 1.0, and the original image size if(myid == main_node): a = get_im(orgstack) nnxo = a.get_xsize() if ali3d_options.CTF: i = a.get_attr('ctf') pixel_size = i.apix else: pixel_size = 1.0 del a else: nnxo = 0 pixel_size = 1.0 pixel_size = bcast_number_to_all(pixel_size, source_node = main_node) nnxo = bcast_number_to_all(nnxo, source_node = main_node) if(radi < 1): radi = nnxo//2-2 elif((2*radi+2)>nnxo): ERROR("HERE","particle radius set too large!",1) ali3d_options.ou = radi if(nxinit < 0): nxinit = min(32, nnxo) nxshrink = nxinit minshrink = 32.0/float(nnxo) shrink = max(float(nxshrink)/float(nnxo),minshrink) # MASTER DIRECTORY if(myid == main_node): print( " masterdir ",masterdir) if( masterdir == ""): timestring = strftime("_%d_%b_%Y_%H_%M_%S", localtime()) masterdir = "master"+timestring li = len(masterdir) cmd = "{} {}".format("mkdir", masterdir) cmdexecute(cmd) keepchecking = 0 else: li = 0 keepchecking = 1 else: li = 0 keepchecking = 1 li = mpi_bcast(li,1,MPI_INT,main_node,MPI_COMM_WORLD)[0] if( li > 0 ): masterdir = mpi_bcast(masterdir,li,MPI_CHAR,main_node,MPI_COMM_WORLD) masterdir = string.join(masterdir,"") # create a vstack from input stack to the local stack in masterdir # Stack name set to default stack = "bdb:"+masterdir+"/rdata" # Initialization of stacks if(myid == main_node): if keepchecking: if(os.path.exists(os.path.join(masterdir,"EMAN2DB/rdata.bdb"))): doit = False else: doit = True else: doit = True if doit: if(orgstack[:4] == "bdb:"): cmd = "{} {} {}".format("e2bdb.py", orgstack,"--makevstack="+stack) else: cmd = "{} {} {}".format("sxcpy.py", orgstack, stack) cmdexecute(cmd) cmd = "{} {}".format("sxheader.py --consecutive --params=originalid", stack) cmdexecute(cmd) keepchecking = False total_stack = EMUtil.get_image_count(stack) junk = get_im(stack) nnxo = junk.get_xsize() del junk else: total_stack = 0 nnxo = 0 total_stack = bcast_number_to_all(total_stack, source_node = main_node) nnxo = bcast_number_to_all(nnxo, source_node = main_node) # INITIALIZATION # Run exhaustive projection matching to get initial orientation parameters # Estimate initial resolution initdir = os.path.join(masterdir,"main000") # make sure the initial volume is not set to zero outside of a mask, as if it is it will crach the program if( myid == main_node and (not options.startangles)): viv = get_im(volinit) if(options.mask3D == None): mask33d = model_circle(radi,nnxo,nnxo,nnxo) else: mask33d = (options.mask3D).copy() st = Util.infomask(viv, mask33d, False) if( st[0] == 0.0 ): viv += (model_blank(nnxo,nnxo,nnxo,1.0) - mask33d)*model_gauss_noise(st[1]/1000.0,nnxo,nnxo,nnxo) viv.write_image(volinit) del mask33d, viv doit, keepchecking = checkstep(initdir, keepchecking, myid, main_node) if doit: partids = os.path.join(masterdir, "ids.txt") partstack = os.path.join(masterdir, "paramszero.txt") xr = min(8,(nnxo - (2*radi+1))//2) if(xr > 3): ts = "2" else: ts = "1" delta = int(options.delta) if(delta <= 0.0): delta = "%f"%round(degrees(atan(1.0/float(radi))), 2) paramsdict = { "stack":stack,"delta":"2.0", "ts":ts, "xr":"%f"%xr, "an":"-1", "center":options.center, "maxit":1, \ "currentres":0.4, "aa":0.1, "radius":radi, "nsoft":0, "delpreviousmax":True, "shrink":1.0, "saturatecrit":1.0, \ "refvol":volinit, "mask3D":options.mask3D} if(options.startangles): if( myid == main_node ): cmd = "mkdir "+initdir cmdexecute(cmd) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"INITIALIZATION") cmd = "{} {}".format("sxheader.py --params=xform.projection --export="+os.path.join(initdir,"params-chunk0.txt"), stack) cmdexecute(cmd) print(line,"Executed successfully: ","Imported initial parameters from the input stack") else: if( myid == main_node ): line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"INITIALIZATION") write_text_file(range(total_stack), partids) write_text_row([[0.0,0.0,0.0,0.0,0.0] for i in xrange(total_stack) ], partstack) metamove(paramsdict, partids, partstack, initdir, 0, myid, main_node, nproc) if(myid == main_node): print(line,"Executed successfully: ","initialization ali3d_base_MPI %d"%nsoft) # store params partids = [None]*2 for procid in xrange(2): partids[procid] = os.path.join(initdir,"chunk%01d.txt"%procid) partstack = [None]*2 for procid in xrange(2): partstack[procid] = os.path.join(initdir,"params-chunk%01d.txt"%procid) from random import shuffle if(myid == main_node): # split randomly params = read_text_row(os.path.join(initdir,"params-chunk0.txt")) assert(len(params) == total_stack) ll = range(total_stack) shuffle(ll) l1 = ll[:total_stack//2] l2 = ll[total_stack//2:] del ll l1.sort() l2.sort() write_text_file(l1,partids[0]) write_text_file(l2,partids[1]) write_text_row([params[i] for i in l1], partstack[0]) write_text_row([params[i] for i in l2], partstack[1]) del params, l1, l2 mpi_barrier(MPI_COMM_WORLD) # Now parallel vol = [None]*2 for procid in xrange(2): projdata = getindexdata(stack, partids[procid], partstack[procid], myid, nproc) if ali3d_options.CTF: vol[procid] = recons3d_4nn_ctf_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) else: vol[procid] = recons3d_4nn_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) del projdata if( myid == main_node): vol[procid].write_image(os.path.join(initdir,"vol%01d.hdf"%procid) ) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print( line,"Generated inivol #%01d "%procid) if(myid == main_node): currentres = get_resolution(vol, radi, nnxo, initdir) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print( line,"Initial resolution %6.4f"%currentres) write_text_file([currentres],os.path.join(initdir,"current_resolution.txt")) else: currentres = 0.0 currentres = bcast_number_to_all(currentres, source_node = main_node) else: if(myid == main_node): currentres = read_text_file(os.path.join(initdir,"current_resolution.txt"))[0] else: currentres = 0.0 currentres = bcast_number_to_all(currentres, source_node = main_node) # set for the first iteration nxshrink = min(max(32, int((currentres+paramsdict["aa"]/2.)*2*nnxo + 0.5)), nnxo) shrink = float(nxshrink)/nnxo tracker = {"previous-resolution":currentres, "movedup":False,"eliminated-outliers":False,\ "previous-nx":nxshrink, "previous-shrink":shrink, "extension":0, "bestsolution":0} previousoutputdir = initdir # MAIN ITERATION mainiteration = 0 keepgoing = 1 while(keepgoing): mainiteration += 1 # prepare output directory mainoutputdir = os.path.join(masterdir,"main%03d"%mainiteration) if(myid == main_node): line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"MAIN ITERATION #",mainiteration, shrink, nxshrink) if keepchecking: if(os.path.exists(mainoutputdir)): doit = 0 print("Directory ",mainoutputdir," exists!") else: doit = 1 keepchecking = False else: doit = 1 if doit: cmd = "{} {}".format("mkdir", mainoutputdir) cmdexecute(cmd) # prepare names of input file names, they are in main directory, # log subdirectories contain outputs from specific refinements partids = [None]*2 for procid in xrange(2): partids[procid] = os.path.join(previousoutputdir,"chunk%01d.txt"%procid) partstack = [None]*2 for procid in xrange(2): partstack[procid] = os.path.join(previousoutputdir,"params-chunk%01d.txt"%procid) mpi_barrier(MPI_COMM_WORLD) #mpi_finalize() #exit() #print("RACING A ",myid) outvol = [os.path.join(previousoutputdir,"vol%01d.hdf"%procid) for procid in xrange(2)] for procid in xrange(2): doit, keepchecking = checkstep(outvol[procid], keepchecking, myid, main_node) if doit: from multi_shc import do_volume projdata = getindexdata(stack, partids[procid], partstack[procid], myid, nproc) if ali3d_options.CTF: vol = recons3d_4nn_ctf_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) else: vol = recons3d_4nn_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) del projdata if( myid == main_node): vol.write_image(outvol[procid]) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print( line,"Generated inivol #%01d "%procid) del vol if(myid == main_node): if keepchecking: procid = 1 if(os.path.join(mainoutputdir,"fusevol%01d.hdf"%procid)): doit = 0 else: doit = 1 keepchecking = False else: doit = 1 if doit: vol = [get_im(outvol[procid]) for procid in xrange(2) ] fq = 0.11 # which part to fuse fuselowf(vol, fq) for procid in xrange(2): vol[procid].write_image(os.path.join(mainoutputdir,"fusevol%01d.hdf"%procid) ) del vol else: doit = 0 mpi_barrier(MPI_COMM_WORLD) doit = bcast_number_to_all(doit, source_node = main_node) # Refine two groups at a current resolution lastring = int(shrink*radi + 0.5) if(lastring < 2): print( line,"ERROR!! lastring too small ", radi, shrink, lastring) break # REFINEMENT # Part "a" SHC for procid in xrange(2): coutdir = os.path.join(mainoutputdir,"loga%01d"%procid) doit, keepchecking = checkstep(coutdir , keepchecking, myid, main_node) paramsdict = { "stack":stack,"delta":"%f"%round(degrees(atan(1.0/lastring)), 2) , "ts":"1", "xr":"2", "an":"-1", "center":options.center, "maxit":1500, \ "currentres":currentres, "aa":0.1, "radius":radi, "nsoft":1, "saturatecrit":0.75, "delpreviousmax":True, "shrink":shrink, \ "refvol":os.path.join(mainoutputdir,"fusevol%01d.hdf"%procid),"mask3D":options.mask3D } if doit: metamove(paramsdict, partids[procid], partstack[procid], coutdir, procid, myid, main_node, nproc) partstack = [None]*2 for procid in xrange(2): partstack[procid] = os.path.join(mainoutputdir, "loga%01d"%procid, "params-chunk%01d.txt"%procid) for procid in xrange(2): outvol = os.path.join(mainoutputdir,"loga%01d"%procid,"shcvol%01d.hdf"%procid) doit, keepchecking = checkstep(outvol, keepchecking, myid, main_node) if doit: from multi_shc import do_volume projdata = getindexdata(stack, partids[procid], partstack[procid], myid, nproc) if ali3d_options.CTF: vol = recons3d_4nn_ctf_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) else: vol = recons3d_4nn_MPI(myid, projdata, symmetry=ali3d_options.sym, npad = 2) del projdata if( myid == main_node): vol.write_image(outvol) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print( line,"Generated shcvol #%01d "%procid) del vol if(myid == main_node): if keepchecking: procid = 1 if(os.path.join(mainoutputdir,"loga%01d"%procid,"fusevol%01d.hdf"%procid) ): doit = 0 else: doit = 1 keepchecking = False else: doit = 1 if doit: vol = [] for procid in xrange(2): vol.append(get_im(os.path.join(mainoutputdir,"loga%01d"%procid,"shcvol%01d.hdf"%procid) )) fq = 0.11 # which part to fuse fuselowf(vol, fq) for procid in xrange(2): vol[procid].write_image( os.path.join(mainoutputdir,"loga%01d"%procid,"fusevol%01d.hdf"%procid) ) del vol else: doit = 0 mpi_barrier(MPI_COMM_WORLD) doit = bcast_number_to_all(doit, source_node = main_node) # Part "b" deterministic partstack = [None]*2 for procid in xrange(2): partstack[procid] = os.path.join(mainoutputdir,"loga%01d"%procid,"params-chunk%01d.txt"%procid) for procid in xrange(2): coutdir = os.path.join(mainoutputdir,"logb%01d"%procid) doit, keepchecking = checkstep(coutdir, keepchecking, myid, main_node) # Run exhaustive to finish up matching paramsdict = { "stack":stack,"delta":"%f"%round(degrees(atan(1.0/lastring)), 2) , "ts":"1", "xr":"2", "an":"-1", "center":options.center, "maxit":10, \ "currentres":currentres, "aa":0.1, "radius":radi, "nsoft":0, "saturatecrit":0.95, "delpreviousmax":True, "shrink":shrink, \ "refvol":os.path.join(mainoutputdir,"loga%01d"%procid,"fusevol%01d.hdf"%procid), "mask3D":options.mask3D } if doit: metamove(paramsdict, partids[procid], partstack[procid], coutdir, procid, myid, main_node, nproc) partstack = [None]*2 for procid in xrange(2): partstack[procid] = os.path.join(mainoutputdir,"logb%01d"%procid,"params-chunk%01d.txt"%procid) # Compute current resolution, store result in main directory doit, keepchecking = checkstep(os.path.join(mainoutputdir,"current_resolution.txt"), keepchecking, myid, main_node) newres = 0.0 if doit: newres = compute_resolution(stack, mainoutputdir, partids, partstack, radi, nnxo, ali3d_options.CTF, myid, main_node, nproc) else: if(myid == main_node): newres = read_text_file( os.path.join(mainoutputdir,"current_resolution.txt") )[0] newres = bcast_number_to_all(newres, source_node = main_node) # Here I have code to generate presentable results. IDs and params have to be merged and stored and an overall volume computed. doit, keepchecking = checkstep(os.path.join(mainoutputdir,"volf.hdf"), keepchecking, myid, main_node) if doit: if( myid == main_node ): pinids = map(int, read_text_file(partids[0]) ) + map(int, read_text_file(partids[1]) ) params = read_text_row(partstack[0]) + read_text_row(partstack[1]) assert(len(pinids) == len(params)) for i in xrange(len(pinids)): pinids[i] = [ pinids[i], params[i] ] del params pinids.sort() write_text_file([pinids[i][0] for i in xrange(len(pinids))], os.path.join(mainoutputdir,"indexes.txt")) write_text_row( [pinids[i][1] for i in xrange(len(pinids))], os.path.join(mainoutputdir,"params.txt")) mpi_barrier(MPI_COMM_WORLD) ali3d_options.fl = newres ali3d_options.ou = radi projdata = getindexdata(stack, os.path.join(mainoutputdir,"indexes.txt"), os.path.join(mainoutputdir,"params.txt"), myid, nproc) volf = do_volume(projdata, ali3d_options, mainiteration, mpi_comm = MPI_COMM_WORLD) if(myid == main_node): volf.write_image(os.path.join(mainoutputdir,"volf.hdf")) mpi_barrier(MPI_COMM_WORLD) #print("RACING X ",myid) if(newres == currentres): for procid in xrange(2): coutdir = os.path.join(mainoutputdir,"logc%01d"%procid) doit, keepchecking = checkstep(coutdir, keepchecking, myid, main_node) if doit: # Do cross-check of the results paramsdict = { "stack":stack,"delta":"%f"%round(degrees(atan(1.0/lastring)), 2) , "ts":"1", "xr":"2", "an":"-1", "center":options.center, "maxit":1, \ "currentres":newres, "aa":0.1, "radius":radi, "nsoft":0, "saturatecrit":0.95, "delpreviousmax":True, "shrink":shrink, \ "refvol":os.path.join(mainoutputdir,"vol%01d.hdf"%(1-procid)), "mask3D":options.mask3D } # The cross-check uses parameters from step "b" to make sure shifts are correct. # As the check is exhaustive, angles are ignored metamove(paramsdict, partids[procid], partstack[procid], coutdir, procid, myid, main_node, nproc) # identify bad apples doit, keepchecking = checkstep(os.path.join(mainoutputdir,"badapples.txt"), keepchecking, myid, main_node) if doit: if(myid == main_node): from utilities import get_symt from pixel_error import max_3D_pixel_error ts = get_symt(ali3d_options.sym) badapples = [] deltaerror = 2.0 total_images_now = 0 for procid in xrange(2): bad = [] ids = map(int,read_text_file( partids[procid] )) total_images_now += len(ids) oldp = read_text_row(partstack[procid]) newp = read_text_row(os.path.join(mainoutputdir,"logc%01d"%procid,"params-chunk%01d.txt"%procid)) for i in xrange(len(ids)): t1 = Transform({"type":"spider","phi":oldp[i][0],"theta":oldp[i][1],"psi":oldp[i][2]}) t1.set_trans(Vec2f(-oldp[i][3]*shrink, -oldp[i][4]*shrink)) t2 = Transform({"type":"spider","phi":newp[i][0],"theta":newp[i][1],"psi":newp[i][2]}) t2.set_trans(Vec2f(-newp[i][3]*shrink, -newp[i][4]*shrink)) if(len(ts) > 1): # only do it if it is not c1 pixel_error = +1.0e23 for kts in ts: ut = t2*kts # we do not care which position minimizes the error pixel_error = min(max_3D_pixel_error(t1, ut, lastring), pixel_error) else: pixel_error = max_3D_pixel_error(t1, t2, lastring) if(pixel_error > deltaerror): bad.append(i) if(len(bad)>0): badapples += [ids[bad[i]] for i in xrange(len(bad))] for i in xrange(len(bad)-1,-1,-1): del oldp[bad[i]],ids[bad[i]] if(len(ids) == 0): ERROR("sxpetite","program diverged, all images have large angular errors, most likely the initial model is badly off",1) else: # This generate new parameters, hopefully to be used as starting ones in the new iteration write_text_file(ids,os.path.join(mainoutputdir,"chunk%01d.txt"%procid)) write_text_row(oldp,os.path.join(mainoutputdir,"params-chunk%01d.txt"%procid)) if(len(badapples)>0): badapples.sort() write_text_file(badapples,os.path.join(mainoutputdir,"badapples.txt")) eli = 100*float(len(badapples))/float(total_images_now) line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"Elimination of outliers: %5.1f percent"%eli ) else: eli = 0.0 del badapples, oldp,ids,bad,newp,ts else: eli =0.0 eli = bcast_number_to_all(eli, source_node = main_node) # This part under MPI if(eli > 0.0): # Compute current resolution depres = compute_resolution(stack, mainoutputdir, \ [os.path.join(mainoutputdir,"chunk%01d.txt"%procid) for procid in xrange(2)], \ [os.path.join(mainoutputdir,"params-chunk%01d.txt"%procid) for procid in xrange(2)], \ radi, nnxo, ali3d_options.CTF, myid, main_node, nproc) depres = bcast_number_to_all(depres, source_node = main_node) if(depres < newres): # elimination of outliers decreased resolution, ignore the effort eliminated_outliers = False else: eliminated_outliers = True newres = depres """ # It does not seem to be needed, as data is there, we just point to the directory for procid in xrange(2): # set pointers to current parameters in main, which are for the reduced set stored above partids[procid] = os.path.join(mainoutputdir,"chunk%01d.txt"%procid partstack[procid] = os.path.join(mainoutputdir,"params-chunk%01d.txt"%procid) """ else: eliminated_outliers = False else: eliminated_outliers = False if(myid == main_node and not eliminated_outliers): for procid in xrange(2): # This is standard path, copy parameters to be used to the main cmd = "{} {} {}".format("cp -p ", partids[procid] , os.path.join(mainoutputdir,"chunk%01d.txt"%procid)) cmdexecute(cmd) cmd = "{} {} {}".format("cp -p ", partstack[procid], os.path.join(mainoutputdir,"params-chunk%01d.txt"%procid)) cmdexecute(cmd) keepgoing = 0 if( newres > currentres or (eliminated_outliers and not tracker["eliminated-outliers"])): if(myid == main_node): print(" Resolution improved, full steam ahead!") if( newres > currentres ): tracker["movedup"] = True else: tracker["movedup"] = False shrink = max(min(2*newres + paramsdict["aa"], 1.0),minshrink) tracker["extension"] = 4 nxshrink = min(int(nnxo*shrink + 0.5) + tracker["extension"],nnxo) tracker["previous-resolution"] = newres currentres = newres tracker["bestsolution"] = mainiteration bestoutputdir = mainoutputdir tracker["eliminated-outliers"] = eliminated_outliers keepgoing = 1 elif(newres < currentres): if(not tracker["movedup"] and tracker["extension"] < 2 and mainiteration > 1): keepgoing = 0 if(myid == main_node): print(" Cannot improve resolution, the best result is in the directory main%03d"%tracker["bestsolution"]) else: if(not tracker["movedup"] and tracker["extension"] > 1 and mainiteration > 1): if(myid == main_node): print(" Resolution decreased. Will decrease target resolution and will fall back on the best so far: main%03d"%tracker["bestsolution"]) bestoutputdir = os.path.join(masterdir,"main%03d"%tracker["bestsolution"]) elif( tracker["movedup"] and tracker["extension"] > 1 and mainiteration > 1): if(myid == main_node): print(" Resolution decreased. Will decrease target resolution and will try starting from previous stage: main%03d"%(mainiteration - 1)) bestoutputdir = os.path.join(masterdir,"main%03d"%(mainiteration-1)) elif( mainiteration == 1): if(myid == main_node): print(" Resolution decreased in the first iteration. It is expected, not to worry") bestoutputdir = mainoutputdir tracker["extension"] += 1 else: # missing something here? if(myid == main_node): print(" Should not be here, ERROR 175!") break mpi_finalize() exit() if( bestoutputdir != mainoutputdir ): # This is the key, we just reset the main to previous, so it will be eventually used as a starting in the next iteration mainoutputdir = bestoutputdir """ # Set data from the main previous best to the current. for procid in xrange(2): partids[procid] = os.path.join(bestoutputdir,"chunk%01d.txt"%procid) partstack[procid] = os.path.join(bestoutputdir,"params-chunk%01d.txt"%procid) """ if(myid == main_node): currentres = read_text_file( os.path.join(bestoutputdir,"current_resolution.txt") )[0] currentres = bcast_number_to_all(currentres, source_node = main_node) shrink = max(min(2*currentres + paramsdict["aa"], 1.0), minshrink) tracker["extension"] -= 1 nxshrink = min(int(nnxo*shrink + 0.5) + tracker["extension"],nnxo) tracker["previous-resolution"] = newres tracker["eliminated-outliers"] = eliminated_outliers tracker["movedup"] = False keepgoing = 1 elif(newres == currentres): if( tracker["extension"] > 0 ): if(myid == main_node): print("The resolution did not improve. This is look ahead move. Let's try to relax slightly and hope for the best") tracker["extension"] -= 1 tracker["movedup"] = False shrink = max(min(2*currentres + paramsdict["aa"], 1.0), minshrink) nxshrink = min(int(nnxo*shrink + 0.5) + tracker["extension"],nnxo) if( tracker["previous-nx"] == nnxo ): keepgoing = 0 else: tracker["previous-resolution"] = newres currentres = newres tracker["eliminated-outliers"] = eliminated_outliers tracker["movedup"] = False keepgoing = 1 else: if(myid == main_node): print("The resolution did not improve.") keepgoing = 0 if( keepgoing == 1 ): if(myid == main_node): print(" New shrink and image dimension :",shrink,nxshrink) """ # It does not look like it is necessary, we just have to point to the directory as the files should be there. # Will continue, so update the params files for procid in xrange(2): # partids ads partstack contain parameters to be used as starting in the next iteration if(not os.path.exists(os.path.join(mainoutputdir,"chunk%01d.txt"%procid))): cmd = "{} {} {}".format("cp -p ", partids[procid] , os.path.join(mainoutputdir,"chunk%01d.txt"%procid)) cmdexecute(cmd) if(not os.path.exists(os.path.join(mainoutputdir,"params-chunk%01d.txt"%procid))): cmd = "{} {} {}".format("cp -p ", partstack[procid], os.path.join(mainoutputdir,"params-chunk%01d.txt"%procid)) cmdexecute(cmd) """ previousoutputdir = mainoutputdir tracker["previous-shrink"] = shrink tracker["previous-nx"] = nxshrink else: if(myid == main_node): print(" Terminating, the best solution is in the directory main%03d"%tracker["bestsolution"]) mpi_barrier(MPI_COMM_WORLD) mpi_finalize()
def main(): from optparse import OptionParser from global_def import SPARXVERSION from EMAN2 import EMData from logger import Logger, BaseLogger_Files import sys, os, time global Tracker, Blockdata from global_def import ERROR progname = os.path.basename(sys.argv[0]) usage = progname + " --output_dir=output_dir --isac_dir=output_dir_of_isac " parser = OptionParser(usage, version=SPARXVERSION) parser.add_option("--pw_adjustment", type ="string", default ='analytical_model', \ help="adjust power spectrum of 2-D averages to an analytic model. Other opions: no_adjustment; bfactor; a text file of 1D rotationally averaged PW") #### Four options for --pw_adjustment: # 1> analytical_model(default); # 2> no_adjustment; # 3> bfactor; # 4> adjust_to_given_pw2(user has to provide a text file that contains 1D rotationally averaged PW) # options in common parser.add_option( "--isac_dir", type="string", default='', help="ISAC run output directory, input directory for this command") parser.add_option( "--output_dir", type="string", default='', help="output directory where computed averages are saved") parser.add_option( "--pixel_size", type="float", default=-1.0, help= "pixel_size of raw images. one can put 1.0 in case of negative stain data" ) parser.add_option( "--fl", type="float", default=-1.0, help= "low pass filter, = -1.0, not applied; =0.0, using FH1 (initial resolution), = 1.0 using FH2 (resolution after local alignment), or user provided value in absolute freqency [0.0:0.5]" ) parser.add_option("--stack", type="string", default="", help="data stack used in ISAC") parser.add_option("--radius", type="int", default=-1, help="radius") parser.add_option("--xr", type="float", default=-1.0, help="local alignment search range") #parser.add_option("--ts", type ="float", default =1.0, help= "local alignment search step") parser.add_option("--fh", type="float", default=-1.0, help="local alignment high frequencies limit") #parser.add_option("--maxit", type ="int", default =5, help= "local alignment iterations") parser.add_option("--navg", type="int", default=1000000, help="number of aveages") parser.add_option("--local_alignment", action="store_true", default=False, help="do local alignment") parser.add_option( "--noctf", action="store_true", default=False, help= "no ctf correction, useful for negative stained data. always ctf for cryo data" ) parser.add_option( "--B_start", type="float", default=45.0, help= "start frequency (Angstrom) of power spectrum for B_factor estimation") parser.add_option( "--Bfactor", type="float", default=-1.0, help= "User defined bactors (e.g. 25.0[A^2]). By default, the program automatically estimates B-factor. " ) (options, args) = parser.parse_args(sys.argv[1:]) adjust_to_analytic_model = False adjust_to_given_pw2 = False B_enhance = False no_adjustment = False if options.pw_adjustment == 'analytical_model': adjust_to_analytic_model = True elif options.pw_adjustment == 'no_adjustment': no_adjustment = True elif options.pw_adjustment == 'bfactor': B_enhance = True else: adjust_to_given_pw2 = True from utilities import get_im, bcast_number_to_all, write_text_file, read_text_file, wrap_mpi_bcast, write_text_row from utilities import cmdexecute from filter import filt_tanl from logger import Logger, BaseLogger_Files import user_functions import string from string import split, atoi, atof import json mpi_init(0, []) nproc = mpi_comm_size(MPI_COMM_WORLD) myid = mpi_comm_rank(MPI_COMM_WORLD) Blockdata = {} # MPI stuff Blockdata["nproc"] = nproc Blockdata["myid"] = myid Blockdata["main_node"] = 0 Blockdata["shared_comm"] = mpi_comm_split_type(MPI_COMM_WORLD, MPI_COMM_TYPE_SHARED, 0, MPI_INFO_NULL) Blockdata["myid_on_node"] = mpi_comm_rank(Blockdata["shared_comm"]) Blockdata["no_of_processes_per_group"] = mpi_comm_size( Blockdata["shared_comm"]) masters_from_groups_vs_everything_else_comm = mpi_comm_split( MPI_COMM_WORLD, Blockdata["main_node"] == Blockdata["myid_on_node"], Blockdata["myid_on_node"]) Blockdata["color"], Blockdata["no_of_groups"], balanced_processor_load_on_nodes = get_colors_and_subsets(Blockdata["main_node"], MPI_COMM_WORLD, Blockdata["myid"], \ Blockdata["shared_comm"], Blockdata["myid_on_node"], masters_from_groups_vs_everything_else_comm) # We need two nodes for processing of volumes Blockdata["node_volume"] = [ Blockdata["no_of_groups"] - 3, Blockdata["no_of_groups"] - 2, Blockdata["no_of_groups"] - 1 ] # For 3D stuff take three last nodes # We need two CPUs for processing of volumes, they are taken to be main CPUs on each volume # We have to send the two myids to all nodes so we can identify main nodes on two selected groups. Blockdata["nodes"] = [Blockdata["node_volume"][0]*Blockdata["no_of_processes_per_group"],Blockdata["node_volume"][1]*Blockdata["no_of_processes_per_group"], \ Blockdata["node_volume"][2]*Blockdata["no_of_processes_per_group"]] # End of Blockdata: sorting requires at least three nodes, and the used number of nodes be integer times of three global_def.BATCH = True global_def.MPI = True if adjust_to_given_pw2: checking_flag = 0 if (Blockdata["myid"] == Blockdata["main_node"]): if not os.path.exists(options.pw_adjustment): checking_flag = 1 checking_flag = bcast_number_to_all(checking_flag, Blockdata["main_node"], MPI_COMM_WORLD) if checking_flag == 1: ERROR("User provided power spectrum does not exist", "sxcompute_isac_avg.py", 1, Blockdata["myid"]) Tracker = {} Constants = {} Constants["isac_dir"] = options.isac_dir Constants["masterdir"] = options.output_dir Constants["pixel_size"] = options.pixel_size Constants["orgstack"] = options.stack Constants["radius"] = options.radius Constants["xrange"] = options.xr Constants["FH"] = options.fh Constants["low_pass_filter"] = options.fl #Constants["maxit"] = options.maxit Constants["navg"] = options.navg Constants["B_start"] = options.B_start Constants["Bfactor"] = options.Bfactor if adjust_to_given_pw2: Constants["modelpw"] = options.pw_adjustment Tracker["constants"] = Constants # ------------------------------------------------------------- # # Create and initialize Tracker dictionary with input options # State Variables #<<<---------------------->>>imported functions<<<--------------------------------------------- #x_range = max(Tracker["constants"]["xrange"], int(1./Tracker["ini_shrink"])+1) #y_range = x_range ####----------------------------------------------------------- # Create Master directory and associated subdirectories line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" if Tracker["constants"]["masterdir"] == Tracker["constants"]["isac_dir"]: masterdir = os.path.join(Tracker["constants"]["isac_dir"], "sharpen") else: masterdir = Tracker["constants"]["masterdir"] if (Blockdata["myid"] == Blockdata["main_node"]): msg = "Postprocessing ISAC 2D averages starts" print(line, "Postprocessing ISAC 2D averages starts") if not masterdir: timestring = strftime("_%d_%b_%Y_%H_%M_%S", localtime()) masterdir = "sharpen_" + Tracker["constants"]["isac_dir"] os.mkdir(masterdir) else: if os.path.exists(masterdir): print("%s already exists" % masterdir) else: os.mkdir(masterdir) subdir_path = os.path.join(masterdir, "ali2d_local_params_avg") if not os.path.exists(subdir_path): os.mkdir(subdir_path) subdir_path = os.path.join(masterdir, "params_avg") if not os.path.exists(subdir_path): os.mkdir(subdir_path) li = len(masterdir) else: li = 0 li = mpi_bcast(li, 1, MPI_INT, Blockdata["main_node"], MPI_COMM_WORLD)[0] masterdir = mpi_bcast(masterdir, li, MPI_CHAR, Blockdata["main_node"], MPI_COMM_WORLD) masterdir = string.join(masterdir, "") Tracker["constants"]["masterdir"] = masterdir log_main = Logger(BaseLogger_Files()) log_main.prefix = Tracker["constants"]["masterdir"] + "/" while not os.path.exists(Tracker["constants"]["masterdir"]): print("Node ", Blockdata["myid"], " waiting...", Tracker["constants"]["masterdir"]) sleep(1) mpi_barrier(MPI_COMM_WORLD) if (Blockdata["myid"] == Blockdata["main_node"]): init_dict = {} print(Tracker["constants"]["isac_dir"]) Tracker["directory"] = os.path.join(Tracker["constants"]["isac_dir"], "2dalignment") core = read_text_row( os.path.join(Tracker["directory"], "initial2Dparams.txt")) for im in range(len(core)): init_dict[im] = core[im] del core else: init_dict = 0 init_dict = wrap_mpi_bcast(init_dict, Blockdata["main_node"], communicator=MPI_COMM_WORLD) ### do_ctf = True if options.noctf: do_ctf = False if (Blockdata["myid"] == Blockdata["main_node"]): if do_ctf: print("CTF correction is on") else: print("CTF correction is off") if options.local_alignment: print("local refinement is on") else: print("local refinement is off") if B_enhance: print("Bfactor is to be applied on averages") elif adjust_to_given_pw2: print("PW of averages is adjusted to a given 1D PW curve") elif adjust_to_analytic_model: print("PW of averages is adjusted to analytical model") else: print("PW of averages is not adjusted") #Tracker["constants"]["orgstack"] = "bdb:"+ os.path.join(Tracker["constants"]["isac_dir"],"../","sparx_stack") image = get_im(Tracker["constants"]["orgstack"], 0) Tracker["constants"]["nnxo"] = image.get_xsize() if Tracker["constants"]["pixel_size"] == -1.0: print( "Pixel size value is not provided by user. extracting it from ctf header entry of the original stack." ) try: ctf_params = image.get_attr("ctf") Tracker["constants"]["pixel_size"] = ctf_params.apix except: ERROR( "Pixel size could not be extracted from the original stack.", "sxcompute_isac_avg.py", 1, Blockdata["myid"]) # action=1 - fatal error, exit ## Now fill in low-pass filter isac_shrink_path = os.path.join(Tracker["constants"]["isac_dir"], "README_shrink_ratio.txt") if not os.path.exists(isac_shrink_path): ERROR( "%s does not exist in the specified ISAC run output directory" % (isac_shrink_path), "sxcompute_isac_avg.py", 1, Blockdata["myid"]) # action=1 - fatal error, exit isac_shrink_file = open(isac_shrink_path, "r") isac_shrink_lines = isac_shrink_file.readlines() isac_shrink_ratio = float( isac_shrink_lines[5] ) # 6th line: shrink ratio (= [target particle radius]/[particle radius]) used in the ISAC run isac_radius = float( isac_shrink_lines[6] ) # 7th line: particle radius at original pixel size used in the ISAC run isac_shrink_file.close() print("Extracted parameter values") print("ISAC shrink ratio : {0}".format(isac_shrink_ratio)) print("ISAC particle radius : {0}".format(isac_radius)) Tracker["ini_shrink"] = isac_shrink_ratio else: Tracker["ini_shrink"] = 0.0 Tracker = wrap_mpi_bcast(Tracker, Blockdata["main_node"], communicator=MPI_COMM_WORLD) #print(Tracker["constants"]["pixel_size"], "pixel_size") x_range = max(Tracker["constants"]["xrange"], int(1. / Tracker["ini_shrink"] + 0.99999)) a_range = y_range = x_range if (Blockdata["myid"] == Blockdata["main_node"]): parameters = read_text_row( os.path.join(Tracker["constants"]["isac_dir"], "all_parameters.txt")) else: parameters = 0 parameters = wrap_mpi_bcast(parameters, Blockdata["main_node"], communicator=MPI_COMM_WORLD) params_dict = {} list_dict = {} #parepare params_dict #navg = min(Tracker["constants"]["navg"]*Blockdata["nproc"], EMUtil.get_image_count(os.path.join(Tracker["constants"]["isac_dir"], "class_averages.hdf"))) navg = min( Tracker["constants"]["navg"], EMUtil.get_image_count( os.path.join(Tracker["constants"]["isac_dir"], "class_averages.hdf"))) global_dict = {} ptl_list = [] memlist = [] if (Blockdata["myid"] == Blockdata["main_node"]): print("Number of averages computed in this run is %d" % navg) for iavg in range(navg): params_of_this_average = [] image = get_im( os.path.join(Tracker["constants"]["isac_dir"], "class_averages.hdf"), iavg) members = sorted(image.get_attr("members")) memlist.append(members) for im in range(len(members)): abs_id = members[im] global_dict[abs_id] = [iavg, im] P = combine_params2( init_dict[abs_id][0], init_dict[abs_id][1], init_dict[abs_id][2], init_dict[abs_id][3], \ parameters[abs_id][0], parameters[abs_id][1]/Tracker["ini_shrink"], parameters[abs_id][2]/Tracker["ini_shrink"], parameters[abs_id][3]) if parameters[abs_id][3] == -1: print( "WARNING: Image #{0} is an unaccounted particle with invalid 2D alignment parameters and should not be the member of any classes. Please check the consitency of input dataset." .format(abs_id) ) # How to check what is wrong about mirror = -1 (Toshio 2018/01/11) params_of_this_average.append([P[0], P[1], P[2], P[3], 1.0]) ptl_list.append(abs_id) params_dict[iavg] = params_of_this_average list_dict[iavg] = members write_text_row( params_of_this_average, os.path.join(Tracker["constants"]["masterdir"], "params_avg", "params_avg_%03d.txt" % iavg)) ptl_list.sort() init_params = [None for im in range(len(ptl_list))] for im in range(len(ptl_list)): init_params[im] = [ptl_list[im]] + params_dict[global_dict[ ptl_list[im]][0]][global_dict[ptl_list[im]][1]] write_text_row( init_params, os.path.join(Tracker["constants"]["masterdir"], "init_isac_params.txt")) else: params_dict = 0 list_dict = 0 memlist = 0 params_dict = wrap_mpi_bcast(params_dict, Blockdata["main_node"], communicator=MPI_COMM_WORLD) list_dict = wrap_mpi_bcast(list_dict, Blockdata["main_node"], communicator=MPI_COMM_WORLD) memlist = wrap_mpi_bcast(memlist, Blockdata["main_node"], communicator=MPI_COMM_WORLD) # Now computing! del init_dict tag_sharpen_avg = 1000 ## always apply low pass filter to B_enhanced images to suppress noise in high frequencies enforced_to_H1 = False if B_enhance: if Tracker["constants"]["low_pass_filter"] == -1.0: enforced_to_H1 = True if navg < Blockdata["nproc"]: # Each CPU do one average ERROR("number of nproc is larger than number of averages", "sxcompute_isac_avg.py", 1, Blockdata["myid"]) else: FH_list = [[0, 0.0, 0.0] for im in range(navg)] image_start, image_end = MPI_start_end(navg, Blockdata["nproc"], Blockdata["myid"]) if Blockdata["myid"] == Blockdata["main_node"]: cpu_dict = {} for iproc in range(Blockdata["nproc"]): local_image_start, local_image_end = MPI_start_end( navg, Blockdata["nproc"], iproc) for im in range(local_image_start, local_image_end): cpu_dict[im] = iproc else: cpu_dict = 0 cpu_dict = wrap_mpi_bcast(cpu_dict, Blockdata["main_node"], communicator=MPI_COMM_WORLD) slist = [None for im in range(navg)] ini_list = [None for im in range(navg)] avg1_list = [None for im in range(navg)] avg2_list = [None for im in range(navg)] plist_dict = {} data_list = [None for im in range(navg)] if Blockdata["myid"] == Blockdata["main_node"]: if B_enhance: print( "Avg ID B-factor FH1(Res before ali) FH2(Res after ali)" ) else: print("Avg ID FH1(Res before ali) FH2(Res after ali)") for iavg in range(image_start, image_end): mlist = EMData.read_images(Tracker["constants"]["orgstack"], list_dict[iavg]) for im in range(len(mlist)): #mlist[im]= get_im(Tracker["constants"]["orgstack"], list_dict[iavg][im]) set_params2D(mlist[im], params_dict[iavg][im], xform="xform.align2d") if options.local_alignment: """ new_average1 = within_group_refinement([mlist[kik] for kik in range(0,len(mlist),2)], maskfile= None, randomize= False, ir=1.0, \ ou=Tracker["constants"]["radius"], rs=1.0, xrng=[x_range], yrng=[y_range], step=[Tracker["constants"]["xstep"]], \ dst=0.0, maxit=Tracker["constants"]["maxit"], FH=max(Tracker["constants"]["FH"], FH1), FF=0.02, method="") new_average2 = within_group_refinement([mlist[kik] for kik in range(1,len(mlist),2)], maskfile= None, randomize= False, ir=1.0, \ ou= Tracker["constants"]["radius"], rs=1.0, xrng=[ x_range], yrng=[y_range], step=[Tracker["constants"]["xstep"]], \ dst=0.0, maxit=Tracker["constants"]["maxit"], FH = max(Tracker["constants"]["FH"], FH1), FF=0.02, method="") new_avg, frc, plist = compute_average(mlist, Tracker["constants"]["radius"], do_ctf) """ new_avg, plist, FH2 = refinement_2d_local( mlist, Tracker["constants"]["radius"], a_range, x_range, y_range, CTF=do_ctf, SNR=1.0e10) plist_dict[iavg] = plist FH1 = -1.0 else: new_avg, frc, plist = compute_average( mlist, Tracker["constants"]["radius"], do_ctf) FH1 = get_optimistic_res(frc) FH2 = -1.0 #write_text_file(frc, os.path.join(Tracker["constants"]["masterdir"], "fsc%03d.txt"%iavg)) FH_list[iavg] = [iavg, FH1, FH2] if B_enhance: new_avg, gb = apply_enhancement( new_avg, Tracker["constants"]["B_start"], Tracker["constants"]["pixel_size"], Tracker["constants"]["Bfactor"]) print(" %6d %6.3f %4.3f %4.3f" % (iavg, gb, FH1, FH2)) elif adjust_to_given_pw2: roo = read_text_file(Tracker["constants"]["modelpw"], -1) roo = roo[0] # always on the first column new_avg = adjust_pw_to_model( new_avg, Tracker["constants"]["pixel_size"], roo) print(" %6d %4.3f %4.3f " % (iavg, FH1, FH2)) elif adjust_to_analytic_model: new_avg = adjust_pw_to_model( new_avg, Tracker["constants"]["pixel_size"], None) print(" %6d %4.3f %4.3f " % (iavg, FH1, FH2)) elif no_adjustment: pass if Tracker["constants"]["low_pass_filter"] != -1.0: if Tracker["constants"]["low_pass_filter"] == 0.0: low_pass_filter = FH1 elif Tracker["constants"]["low_pass_filter"] == 1.0: low_pass_filter = FH2 if not options.local_alignment: low_pass_filter = FH1 else: low_pass_filter = Tracker["constants"]["low_pass_filter"] if low_pass_filter >= 0.45: low_pass_filter = 0.45 new_avg = filt_tanl(new_avg, low_pass_filter, 0.02) else: # No low pass filter but if enforced if enforced_to_H1: new_avg = filt_tanl(new_avg, FH1, 0.02) if B_enhance: new_avg = fft(new_avg) new_avg.set_attr("members", list_dict[iavg]) new_avg.set_attr("n_objects", len(list_dict[iavg])) slist[iavg] = new_avg print( strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>", "Refined average %7d" % iavg) ## send to main node to write mpi_barrier(MPI_COMM_WORLD) for im in range(navg): # avg if cpu_dict[im] == Blockdata[ "myid"] and Blockdata["myid"] != Blockdata["main_node"]: send_EMData(slist[im], Blockdata["main_node"], tag_sharpen_avg) elif cpu_dict[im] == Blockdata["myid"] and Blockdata[ "myid"] == Blockdata["main_node"]: slist[im].set_attr("members", memlist[im]) slist[im].set_attr("n_objects", len(memlist[im])) slist[im].write_image( os.path.join(Tracker["constants"]["masterdir"], "class_averages.hdf"), im) elif cpu_dict[im] != Blockdata["myid"] and Blockdata[ "myid"] == Blockdata["main_node"]: new_avg_other_cpu = recv_EMData(cpu_dict[im], tag_sharpen_avg) new_avg_other_cpu.set_attr("members", memlist[im]) new_avg_other_cpu.set_attr("n_objects", len(memlist[im])) new_avg_other_cpu.write_image( os.path.join(Tracker["constants"]["masterdir"], "class_averages.hdf"), im) if options.local_alignment: if cpu_dict[im] == Blockdata["myid"]: write_text_row( plist_dict[im], os.path.join(Tracker["constants"]["masterdir"], "ali2d_local_params_avg", "ali2d_local_params_avg_%03d.txt" % im)) if cpu_dict[im] == Blockdata[ "myid"] and cpu_dict[im] != Blockdata["main_node"]: wrap_mpi_send(plist_dict[im], Blockdata["main_node"], MPI_COMM_WORLD) wrap_mpi_send(FH_list, Blockdata["main_node"], MPI_COMM_WORLD) elif cpu_dict[im] != Blockdata["main_node"] and Blockdata[ "myid"] == Blockdata["main_node"]: dummy = wrap_mpi_recv(cpu_dict[im], MPI_COMM_WORLD) plist_dict[im] = dummy dummy = wrap_mpi_recv(cpu_dict[im], MPI_COMM_WORLD) FH_list[im] = dummy[im] else: if cpu_dict[im] == Blockdata[ "myid"] and cpu_dict[im] != Blockdata["main_node"]: wrap_mpi_send(FH_list, Blockdata["main_node"], MPI_COMM_WORLD) elif cpu_dict[im] != Blockdata["main_node"] and Blockdata[ "myid"] == Blockdata["main_node"]: dummy = wrap_mpi_recv(cpu_dict[im], MPI_COMM_WORLD) FH_list[im] = dummy[im] mpi_barrier(MPI_COMM_WORLD) mpi_barrier(MPI_COMM_WORLD) if options.local_alignment: if Blockdata["myid"] == Blockdata["main_node"]: ali3d_local_params = [None for im in range(len(ptl_list))] for im in range(len(ptl_list)): ali3d_local_params[im] = [ptl_list[im]] + plist_dict[ global_dict[ptl_list[im]][0]][global_dict[ptl_list[im]][1]] write_text_row( ali3d_local_params, os.path.join(Tracker["constants"]["masterdir"], "ali2d_local_params.txt")) write_text_row( FH_list, os.path.join(Tracker["constants"]["masterdir"], "FH_list.txt")) else: if Blockdata["myid"] == Blockdata["main_node"]: write_text_row( FH_list, os.path.join(Tracker["constants"]["masterdir"], "FH_list.txt")) mpi_barrier(MPI_COMM_WORLD) target_xr = 3 target_yr = 3 if (Blockdata["myid"] == 0): cmd = "{} {} {} {} {} {} {} {} {} {}".format("sxchains.py", os.path.join(Tracker["constants"]["masterdir"],"class_averages.hdf"),\ os.path.join(Tracker["constants"]["masterdir"],"junk.hdf"),os.path.join(Tracker["constants"]["masterdir"],"ordered_class_averages.hdf"),\ "--circular","--radius=%d"%Tracker["constants"]["radius"] , "--xr=%d"%(target_xr+1),"--yr=%d"%(target_yr+1),"--align", ">/dev/null") junk = cmdexecute(cmd) cmd = "{} {}".format( "rm -rf", os.path.join(Tracker["constants"]["masterdir"], "junk.hdf")) junk = cmdexecute(cmd) from mpi import mpi_finalize mpi_finalize() exit()
def main(): from utilities import write_text_row, drop_image, model_gauss_noise, get_im, set_params_proj, wrap_mpi_bcast, model_circle import user_functions from applications import MPI_start_end from optparse import OptionParser from global_def import SPARXVERSION from EMAN2 import EMData from multi_shc import multi_shc, do_volume from logger import Logger, BaseLogger_Files import sys import os import time import socket progname = os.path.basename(sys.argv[0]) usage = progname + " stack [output_directory] initial_volume --ir=inner_radius --ou=outer_radius --rs=ring_step --xr=x_range --yr=y_range --ts=translational_search_step --delta=angular_step --an=angular_neighborhood --CTF --fl --aa --ref_a=S --sym=c1" parser = OptionParser(usage,version=SPARXVERSION) parser.add_option("--ir", type= "int", default= 1, help="inner radius for rotational correlation > 0 (set to 1)") parser.add_option("--ou", type= "int", default= -1, help="outer radius for rotational correlation < int(nx/2)-1 (set to the radius of the particle)") parser.add_option("--rs", type= "int", default= 1, help="step between rings in rotational correlation >0 (set to 1)" ) parser.add_option("--xr", type="string", default= "-1", help="range for translation search in x direction, search is +/xr (default 0)") parser.add_option("--yr", type="string", default= "-1", help="range for translation search in y direction, search is +/yr (default = same as xr)") parser.add_option("--ts", type="string", default= "1", help="step size of the translation search in both directions, search is -xr, -xr+ts, 0, xr-ts, xr, can be fractional") parser.add_option("--delta", type="string", default= "-1", help="angular step of reference projections during initialization step (default automatically selected based on radius of the structure.)") parser.add_option("--an", type="string", default= "-1", help="angular neighborhood for local searches (phi and theta) (Default exhaustive searches)") parser.add_option("--CTF", action="store_true", default=False, help="Use CTF (Default no CTF correction)") parser.add_option("--shrink", type="float", default= 1.0, help="Reduce data size by shrink factor (default 1.0)") parser.add_option("--snr", type="float", default= 1.0, help="Signal-to-Noise Ratio of the data (default 1.0)") parser.add_option("--ref_a", type="string", default= "S", help="method for generating the quasi-uniformly distributed projection directions (default S)") parser.add_option("--sym", type="string", default= "c1", help="symmetry of the refined structure") parser.add_option("--npad", type="int", default= 2, help="padding size for 3D reconstruction (default=2)") #options introduced for the do_volume function parser.add_option("--fl", type="float", default=0.12, help="cut-off frequency of hyperbolic tangent low-pass Fourier filte (default 0.12)") parser.add_option("--aa", type="float", default=0.1, help="fall-off of hyperbolic tangent low-pass Fourier filter (default 0.1)") parser.add_option("--pwreference", type="string", default="", help="text file with a reference power spectrum (default no power spectrum adjustment)") parser.add_option("--mask3D", type="string", default=None, help="3D mask file (default a sphere WHAT RADIUS??)") (options, args) = parser.parse_args(sys.argv[1:]) #print( " args ",args) if( len(args) == 3): volinit = args[2] masterdir = args[1] elif(len(args) == 2): volinit = args[1] masterdir = "" else: print( "usage: " + usage) print( "Please run '" + progname + " -h' for detailed options") return 1 stack = args[0] # INPUT PARAMETERS radi = options.ou global_def.BATCH = True ali3d_options.ir = options.ir ali3d_options.rs = options.rs ali3d_options.ou = options.ou ali3d_options.xr = options.xr ali3d_options.yr = options.yr ali3d_options.ts = options.ts ali3d_options.an = "-1" ali3d_options.sym = options.sym ali3d_options.delta = options.delta ali3d_options.npad = options.npad ali3d_options.CTF = options.CTF ali3d_options.ref_a = options.ref_a ali3d_options.snr = options.snr ali3d_options.mask3D = options.mask3D ali3d_options.pwreference = "" # It will have to be turned on after exhaustive done by setting to options.pwreference ali3d_options.fl = 0.4 ali3d_options.initfl = 0.4 ali3d_options.aa = 0.1 mpi_init(0, []) nproc = mpi_comm_size(MPI_COMM_WORLD) myid = mpi_comm_rank(MPI_COMM_WORLD) main_node = 0 # Get the pixel size, if none set to 1.0, and the original image size if(myid == main_node): total_stack = EMUtil.get_image_count(stack) a = get_im(stack) nxinit = a.get_xsize() if ali3d_options.CTF: i = a.get_attr('ctf') pixel_size = i.apix fq = pixel_size/fq else: pixel_size = 1.0 # No pixel size, fusing computed as 5 Fourier pixels fq = 5.0/nxinit del a else: total_stack = 0 nxinit = 0 pixel_size = 1.0 total_stack = bcast_number_to_all(total_stack, source_node = main_node) pixel_size = bcast_number_to_all(pixel_size, source_node = main_node) nxinit = bcast_number_to_all(nxinit, source_node = main_node) if(radi < 1): radi = nxinit//2-2 elif((2*radi+2)>nxinit): ERROR("Particle radius set too large!","sxcenter_projections",1,myid) ali3d_options.ou = radi shrink = options.shrink nxshrink = int(nxinit*shrink+0.5) angular_neighborhood = "-1" # MASTER DIRECTORY if(myid == main_node): print( " masterdir ",masterdir) if( masterdir == ""): timestring = strftime("_%d_%b_%Y_%H_%M_%S", localtime()) masterdir = "master"+timestring li = len(masterdir) cmd = "{} {}".format("mkdir", masterdir) junk = cmdexecute(cmd) else: li = 0 li = mpi_bcast(li,1,MPI_INT,main_node,MPI_COMM_WORLD)[0] if( li > 0 ): masterdir = mpi_bcast(masterdir,li,MPI_CHAR,main_node,MPI_COMM_WORLD) masterdir = string.join(masterdir,"") nnxo = nxinit # INITIALIZATION initdir = masterdir # This is initial setting, has to be initialized here, we do not want it to run too long. # INITIALIZATION THAT FOLLOWS WILL HAVE TO BE CHANGED SO THE USER CAN PROVIDE INITIAL GUESS OF RESOLUTION # If we new the initial resolution, it could be done more densely if(options.xr == "-1"): xr = "%d"%((nnxo - (2*radi-1))//2) else: xr = options.xr if(options.yr == "-1"): yr = xr else: yr = options.yr delta = float(options.delta) if(delta <= 0.0): delta = "%f"%round(degrees(atan(1.0/float(radi))), 2) else: delta = "%f"%delta paramsdict = { "stack":stack,"delta":delta, "ts":"1.0", "xr":xr, "an":angular_neighborhood, \ "center":"0", "maxit":1, "local":False,\ "lowpass":options.fl, "initialfl":0.4, "falloff":options.aa, "radius":radi, \ "nsoft":0, "delpreviousmax":True, "shrink":options.shrink, "saturatecrit":1.0, "pixercutoff":2.0,\ "refvol":volinit, "mask3D":options.mask3D} partids = os.path.join(masterdir, "ids.txt") partstack = os.path.join(masterdir, "paramszero.txt") if( myid == main_node ): write_text_file(range(total_stack), partids) write_text_row([[0.0,0.0,0.0,0.0,0.0] for i in xrange(total_stack) ], partstack) run3Dalignment(paramsdict, partids, partstack, initdir, 0, myid, main_node, nproc) mpi_barrier(MPI_COMM_WORLD) mpi_finalize()
def run3Dalignment(paramsdict, partids, partstack, outputdir, procid, myid, main_node, nproc): # Reads from paramsdict["stack"] particles partids set parameters in partstack # and do refinement as specified in paramsdict # # Will create outputdir # Will write to outputdir output parameters: params-chunk0.txt and params-chunk1.txt if(myid == main_node): # Create output directory log = Logger(BaseLogger_Files()) log.prefix = os.path.join(outputdir) #cmd = "mkdir "+log.prefix #junk = cmdexecute(cmd) log.prefix += "/" else: log = None mpi_barrier(MPI_COMM_WORLD) ali3d_options.delta = paramsdict["delta"] ali3d_options.ts = paramsdict["ts"] ali3d_options.xr = paramsdict["xr"] # low pass filter is applied to shrank data, so it has to be adjusted ali3d_options.fl = paramsdict["lowpass"]/paramsdict["shrink"] ali3d_options.initfl = paramsdict["initialfl"]/paramsdict["shrink"] ali3d_options.aa = paramsdict["falloff"] ali3d_options.maxit = paramsdict["maxit"] ali3d_options.mask3D = paramsdict["mask3D"] ali3d_options.an = paramsdict["an"] ali3d_options.ou = paramsdict["radius"] # This is changed in ali3d_base, but the shrank value is needed in vol recons, fixt it! shrinkage = paramsdict["shrink"] projdata = getindexdata(paramsdict["stack"], partids, partstack, myid, nproc) onx = projdata[0].get_xsize() last_ring = ali3d_options.ou if last_ring < 0: last_ring = int(onx/2) - 2 mask2D = model_circle(last_ring,onx,onx) - model_circle(ali3d_options.ir,onx,onx) if(shrinkage < 1.0): # get the new size masks2D = resample(mask2D, shrinkage) nx = masks2D.get_xsize() masks2D = model_circle(int(last_ring*shrinkage+0.5),nx,nx) - model_circle(max(int(ali3d_options.ir*shrinkage+0.5),1),nx,nx) nima = len(projdata) oldshifts = [0.0,0.0]*nima for im in xrange(nima): #data[im].set_attr('ID', list_of_particles[im]) ctf_applied = projdata[im].get_attr_default('ctf_applied', 0) phi,theta,psi,sx,sy = get_params_proj(projdata[im]) projdata[im] = fshift(projdata[im], sx, sy) set_params_proj(projdata[im],[phi,theta,psi,0.0,0.0]) # For local SHC set anchor #if(nsoft == 1 and an[0] > -1): # set_params_proj(data[im],[phi,tetha,psi,0.0,0.0], "xform.anchor") oldshifts[im] = [sx,sy] if ali3d_options.CTF : ctf_params = projdata[im].get_attr("ctf") if ctf_applied == 0: st = Util.infomask(projdata[im], mask2D, False) projdata[im] -= st[0] projdata[im] = filt_ctf(projdata[im], ctf_params) projdata[im].set_attr('ctf_applied', 1) if(shrinkage < 1.0): #phi,theta,psi,sx,sy = get_params_proj(projdata[im]) projdata[im] = resample(projdata[im], shrinkage) st = Util.infomask(projdata[im], None, True) projdata[im] -= st[0] st = Util.infomask(projdata[im], masks2D, True) projdata[im] /= st[1] #sx *= shrinkage #sy *= shrinkage #set_params_proj(projdata[im], [phi,theta,psi,sx,sy]) if ali3d_options.CTF : ctf_params.apix /= shrinkage projdata[im].set_attr('ctf', ctf_params) else: st = Util.infomask(projdata[im], None, True) projdata[im] -= st[0] st = Util.infomask(projdata[im], mask2D, True) projdata[im] /= st[1] del mask2D if(shrinkage < 1.0): del masks2D """ if(paramsdict["delpreviousmax"]): for i in xrange(len(projdata)): try: projdata[i].del_attr("previousmax") except: pass """ if(myid == main_node): print_dict(paramsdict,"3D alignment parameters") print(" => actual lowpass : "******" => actual init lowpass : "******" => PW adjustment : ",ali3d_options.pwreference) print(" => partids : ",partids) print(" => partstack : ",partstack) if(ali3d_options.fl > 0.46): ERROR("Low pass filter in 3D alignment > 0.46 on the scale of shrank data","sxcenter_projections",1,myid) # Run alignment command, it returns params per CPU params = center_projections_3D(projdata, paramsdict["refvol"], \ ali3d_options, onx, shrinkage, \ mpi_comm = MPI_COMM_WORLD, myid = myid, main_node = main_node, log = log ) del log, projdata params = wrap_mpi_gatherv(params, main_node, MPI_COMM_WORLD) # store params if(myid == main_node): for im in xrange(nima): params[im][0] = params[im][0]/shrinkage +oldshifts[im][0] params[im][1] = params[im][1]/shrinkage +oldshifts[im][1] line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"Executed successfully: ","3D alignment"," number of images:%7d"%len(params)) write_text_row(params, os.path.join(outputdir,"params.txt") )
def main(): from utilities import write_text_row, drop_image, model_gauss_noise, get_im, set_params_proj, wrap_mpi_bcast, model_circle import user_functions from applications import MPI_start_end from optparse import OptionParser from global_def import SPARXVERSION from EMAN2 import EMData from multi_shc import multi_shc, do_volume from logger import Logger, BaseLogger_Files import sys import os import time import socket progname = os.path.basename(sys.argv[0]) usage = progname + " stack [output_directory] initial_volume --ir=inner_radius --ou=outer_radius --rs=ring_step --xr=x_range --yr=y_range --ts=translational_search_step --delta=angular_step --an=angular_neighborhood --CTF --fl --aa --ref_a=S --sym=c1" parser = OptionParser(usage,version=SPARXVERSION) parser.add_option("--ir", type= "int", default= 1, help="inner radius for rotational correlation > 0 (set to 1)") parser.add_option("--ou", type= "int", default= -1, help="outer radius for rotational correlation < int(nx/2)-1 (set to the radius of the particle)") parser.add_option("--rs", type= "int", default= 1, help="step between rings in rotational correlation >0 (set to 1)" ) parser.add_option("--xr", type="string", default= "-1", help="range for translation search in x direction, search is +/xr (default 0)") parser.add_option("--yr", type="string", default= "-1", help="range for translation search in y direction, search is +/yr (default = same as xr)") parser.add_option("--ts", type="string", default= "1", help="step size of the translation search in both directions, search is -xr, -xr+ts, 0, xr-ts, xr, can be fractional") parser.add_option("--delta", type="string", default= "-1", help="angular step of reference projections during initialization step (default automatically selected based on radius of the structure.)") parser.add_option("--an", type="string", default= "-1", help="angular neighborhood for local searches (phi and theta) (Default exhaustive searches)") parser.add_option("--CTF", action="store_true", default=False, help="Use CTF (Default no CTF correction)") parser.add_option("--shrink", type="float", default= 1.0, help="Reduce data size by shrink factor (default 1.0)") parser.add_option("--snr", type="float", default= 1.0, help="Signal-to-Noise Ratio of the data (default 1.0)") parser.add_option("--ref_a", type="string", default= "S", help="method for generating the quasi-uniformly distributed projection directions (default S)") parser.add_option("--sym", type="string", default= "c1", help="symmetry of the refined structure") parser.add_option("--npad", type="int", default= 2, help="padding size for 3D reconstruction (default=2)") #options introduced for the do_volume function parser.add_option("--fl", type="float", default=0.12, help="cut-off frequency of hyperbolic tangent low-pass Fourier filte (default 0.12)") parser.add_option("--aa", type="float", default=0.1, help="fall-off of hyperbolic tangent low-pass Fourier filter (default 0.1)") parser.add_option("--pwreference", type="string", default="", help="text file with a reference power spectrum (default no power spectrum adjustment)") parser.add_option("--mask3D", type="string", default=None, help="3D mask file (default a sphere WHAT RADIUS??)") (options, args) = parser.parse_args(sys.argv[1:]) #print( " args ",args) if( len(args) == 3): volinit = args[2] masterdir = args[1] elif(len(args) == 2): volinit = args[1] masterdir = "" else: print( "usage: " + usage) print( "Please run '" + progname + " -h' for detailed options") return 1 stack = args[0] # INPUT PARAMETERS radi = options.ou global_def.BATCH = True ali3d_options.ir = options.ir ali3d_options.rs = options.rs ali3d_options.ou = options.ou ali3d_options.xr = options.xr ali3d_options.yr = options.yr ali3d_options.ts = options.ts ali3d_options.an = "-1" ali3d_options.sym = options.sym ali3d_options.delta = options.delta ali3d_options.npad = options.npad ali3d_options.CTF = options.CTF ali3d_options.ref_a = options.ref_a ali3d_options.snr = options.snr ali3d_options.mask3D = options.mask3D ali3d_options.pwreference = "" # It will have to be turned on after exhaustive done by setting to options.pwreference ali3d_options.fl = 0.4 ali3d_options.initfl = 0.4 ali3d_options.aa = 0.1 mpi_init(0, []) nproc = mpi_comm_size(MPI_COMM_WORLD) myid = mpi_comm_rank(MPI_COMM_WORLD) main_node = 0 # Get the pixel size, if none set to 1.0, and the original image size if(myid == main_node): total_stack = EMUtil.get_image_count(stack) a = get_im(stack) nxinit = a.get_xsize() if ali3d_options.CTF: i = a.get_attr('ctf') pixel_size = i.apix fq = pixel_size/fq else: pixel_size = 1.0 # No pixel size, fusing computed as 5 Fourier pixels fq = 5.0/nxinit del a else: total_stack = 0 nxinit = 0 pixel_size = 1.0 total_stack = bcast_number_to_all(total_stack, source_node = main_node) pixel_size = bcast_number_to_all(pixel_size, source_node = main_node) nxinit = bcast_number_to_all(nxinit, source_node = main_node) if(radi < 1): radi = nxinit//2-2 elif((2*radi+2)>nxinit): ERROR("Particle radius set too large!","sxcenter_projections",1,myid) ali3d_options.ou = radi shrink = options.shrink nxshrink = int(nxinit*shrink+0.5) angular_neighborhood = "-1" # MASTER DIRECTORY if(myid == main_node): print( " masterdir ",masterdir) if( masterdir == ""): timestring = strftime("_%d_%b_%Y_%H_%M_%S", localtime()) masterdir = "master"+timestring li = len(masterdir) cmd = "{} {}".format("mkdir", masterdir) cmdexecute(cmd) else: li = 0 li = mpi_bcast(li,1,MPI_INT,main_node,MPI_COMM_WORLD)[0] if( li > 0 ): masterdir = mpi_bcast(masterdir,li,MPI_CHAR,main_node,MPI_COMM_WORLD) masterdir = string.join(masterdir,"") nnxo = nxinit # INITIALIZATION initdir = masterdir # This is initial setting, has to be initialized here, we do not want it to run too long. # INITIALIZATION THAT FOLLOWS WILL HAVE TO BE CHANGED SO THE USER CAN PROVIDE INITIAL GUESS OF RESOLUTION # If we new the initial resolution, it could be done more densely if(options.xr == "-1"): xr = "%d"%((nnxo - (2*radi-1))//2) else: xr = options.xr if(options.yr == "-1"): yr = xr else: yr = options.yr delta = float(options.delta) if(delta <= 0.0): delta = "%f"%round(degrees(atan(1.0/float(radi))), 2) else: delta = "%f"%delta paramsdict = { "stack":stack,"delta":delta, "ts":"1.0", "xr":xr, "an":angular_neighborhood, \ "center":"0", "maxit":1, "local":False,\ "lowpass":options.fl, "initialfl":0.4, "falloff":options.aa, "radius":radi, \ "nsoft":0, "delpreviousmax":True, "shrink":options.shrink, "saturatecrit":1.0, "pixercutoff":2.0,\ "refvol":volinit, "mask3D":options.mask3D} partids = os.path.join(masterdir, "ids.txt") partstack = os.path.join(masterdir, "paramszero.txt") if( myid == main_node ): write_text_file(range(total_stack), partids) write_text_row([[0.0,0.0,0.0,0.0,0.0] for i in xrange(total_stack) ], partstack) run3Dalignment(paramsdict, partids, partstack, initdir, 0, myid, main_node, nproc) mpi_barrier(MPI_COMM_WORLD) mpi_finalize()
def run3Dalignment(paramsdict, partids, partstack, outputdir, procid, myid, main_node, nproc): # Reads from paramsdict["stack"] particles partids set parameters in partstack # and do refinement as specified in paramsdict # # Will create outputdir # Will write to outputdir output parameters: params-chunk0.txt and params-chunk1.txt if(myid == main_node): # Create output directory log = Logger(BaseLogger_Files()) log.prefix = os.path.join(outputdir) #cmd = "mkdir "+log.prefix #cmdexecute(cmd) log.prefix += "/" else: log = None mpi_barrier(MPI_COMM_WORLD) ali3d_options.delta = paramsdict["delta"] ali3d_options.ts = paramsdict["ts"] ali3d_options.xr = paramsdict["xr"] # low pass filter is applied to shrank data, so it has to be adjusted ali3d_options.fl = paramsdict["lowpass"]/paramsdict["shrink"] ali3d_options.initfl = paramsdict["initialfl"]/paramsdict["shrink"] ali3d_options.aa = paramsdict["falloff"] ali3d_options.maxit = paramsdict["maxit"] ali3d_options.mask3D = paramsdict["mask3D"] ali3d_options.an = paramsdict["an"] ali3d_options.ou = paramsdict["radius"] # This is changed in ali3d_base, but the shrank value is needed in vol recons, fixt it! shrinkage = paramsdict["shrink"] projdata = getindexdata(paramsdict["stack"], partids, partstack, myid, nproc) onx = projdata[0].get_xsize() last_ring = ali3d_options.ou if last_ring < 0: last_ring = int(onx/2) - 2 mask2D = model_circle(last_ring,onx,onx) - model_circle(ali3d_options.ir,onx,onx) if(shrinkage < 1.0): # get the new size masks2D = resample(mask2D, shrinkage) nx = masks2D.get_xsize() masks2D = model_circle(int(last_ring*shrinkage+0.5),nx,nx) - model_circle(max(int(ali3d_options.ir*shrinkage+0.5),1),nx,nx) nima = len(projdata) oldshifts = [0.0,0.0]*nima for im in xrange(nima): #data[im].set_attr('ID', list_of_particles[im]) ctf_applied = projdata[im].get_attr_default('ctf_applied', 0) phi,theta,psi,sx,sy = get_params_proj(projdata[im]) projdata[im] = fshift(projdata[im], sx, sy) set_params_proj(projdata[im],[phi,theta,psi,0.0,0.0]) # For local SHC set anchor #if(nsoft == 1 and an[0] > -1): # set_params_proj(data[im],[phi,tetha,psi,0.0,0.0], "xform.anchor") oldshifts[im] = [sx,sy] if ali3d_options.CTF : ctf_params = projdata[im].get_attr("ctf") if ctf_applied == 0: st = Util.infomask(projdata[im], mask2D, False) projdata[im] -= st[0] projdata[im] = filt_ctf(projdata[im], ctf_params) projdata[im].set_attr('ctf_applied', 1) if(shrinkage < 1.0): #phi,theta,psi,sx,sy = get_params_proj(projdata[im]) projdata[im] = resample(projdata[im], shrinkage) st = Util.infomask(projdata[im], None, True) projdata[im] -= st[0] st = Util.infomask(projdata[im], masks2D, True) projdata[im] /= st[1] #sx *= shrinkage #sy *= shrinkage #set_params_proj(projdata[im], [phi,theta,psi,sx,sy]) if ali3d_options.CTF : ctf_params.apix /= shrinkage projdata[im].set_attr('ctf', ctf_params) else: st = Util.infomask(projdata[im], None, True) projdata[im] -= st[0] st = Util.infomask(projdata[im], mask2D, True) projdata[im] /= st[1] del mask2D if(shrinkage < 1.0): del masks2D """ if(paramsdict["delpreviousmax"]): for i in xrange(len(projdata)): try: projdata[i].del_attr("previousmax") except: pass """ if(myid == main_node): print_dict(paramsdict,"3D alignment parameters") print(" => actual lowpass : "******" => actual init lowpass : "******" => PW adjustment : ",ali3d_options.pwreference) print(" => partids : ",partids) print(" => partstack : ",partstack) if(ali3d_options.fl > 0.46): ERROR("Low pass filter in 3D alignment > 0.46 on the scale of shrank data","sxcenter_projections",1,myid) # Run alignment command, it returns params per CPU params = center_projections_3D(projdata, paramsdict["refvol"], \ ali3d_options, onx, shrinkage, \ mpi_comm = MPI_COMM_WORLD, myid = myid, main_node = main_node, log = log ) del log, projdata params = wrap_mpi_gatherv(params, main_node, MPI_COMM_WORLD) # store params if(myid == main_node): for im in xrange(nima): params[im][0] = params[im][0]/shrinkage +oldshifts[im][0] params[im][1] = params[im][1]/shrinkage +oldshifts[im][1] line = strftime("%Y-%m-%d_%H:%M:%S", localtime()) + " =>" print(line,"Executed successfully: ","3D alignment"," number of images:%7d"%len(params)) write_text_row(params, os.path.join(outputdir,"params.txt") )
def main(): def params_3D_2D_NEW(phi, theta, psi, s2x, s2y, mirror): # the final ali2d parameters already combine shifts operation first and rotation operation second for parameters converted from 3D if mirror: m = 1 alpha, sx, sy, scalen = compose_transform2(0, s2x, s2y, 1.0, 540.0-psi, 0, 0, 1.0) else: m = 0 alpha, sx, sy, scalen = compose_transform2(0, s2x, s2y, 1.0, 360.0-psi, 0, 0, 1.0) return alpha, sx, sy, m progname = os.path.basename(sys.argv[0]) usage = progname + " prj_stack --ave2D= --var2D= --ave3D= --var3D= --img_per_grp= --fl= --aa= --sym=symmetry --CTF" parser = OptionParser(usage, version=SPARXVERSION) parser.add_option("--output_dir", type="string" , default="./", help="Output directory") parser.add_option("--ave2D", type="string" , default=False, help="Write to the disk a stack of 2D averages") parser.add_option("--var2D", type="string" , default=False, help="Write to the disk a stack of 2D variances") parser.add_option("--ave3D", type="string" , default=False, help="Write to the disk reconstructed 3D average") parser.add_option("--var3D", type="string" , default=False, help="Compute 3D variability (time consuming!)") parser.add_option("--img_per_grp", type="int" , default=100, help="Number of neighbouring projections.(Default is 100)") parser.add_option("--no_norm", action="store_true", default=False, help="Do not use normalization.(Default is to apply normalization)") #parser.add_option("--radius", type="int" , default=-1 , help="radius for 3D variability" ) parser.add_option("--npad", type="int" , default=2 , help="Number of time to pad the original images.(Default is 2 times padding)") parser.add_option("--sym" , type="string" , default="c1", help="Symmetry. (Default is no symmetry)") parser.add_option("--fl", type="float" , default=0.0, help="Low pass filter cutoff in absolute frequency (0.0 - 0.5) and is applied to decimated images. (Default - no filtration)") parser.add_option("--aa", type="float" , default=0.02 , help="Fall off of the filter. Use default value if user has no clue about falloff (Default value is 0.02)") parser.add_option("--CTF", action="store_true", default=False, help="Use CFT correction.(Default is no CTF correction)") #parser.add_option("--MPI" , action="store_true", default=False, help="use MPI version") #parser.add_option("--radiuspca", type="int" , default=-1 , help="radius for PCA" ) #parser.add_option("--iter", type="int" , default=40 , help="maximum number of iterations (stop criterion of reconstruction process)" ) #parser.add_option("--abs", type="float" , default=0.0 , help="minimum average absolute change of voxels' values (stop criterion of reconstruction process)" ) #parser.add_option("--squ", type="float" , default=0.0 , help="minimum average squared change of voxels' values (stop criterion of reconstruction process)" ) parser.add_option("--VAR" , action="store_true", default=False, help="Stack of input consists of 2D variances (Default False)") parser.add_option("--decimate", type ="float", default=0.25, help="Image decimate rate, a number less than 1. (Default is 0.25)") parser.add_option("--window", type ="int", default=0, help="Target image size relative to original image size. (Default value is zero.)") #parser.add_option("--SND", action="store_true", default=False, help="compute squared normalized differences (Default False)") #parser.add_option("--nvec", type="int" , default=0 , help="Number of eigenvectors, (Default = 0 meaning no PCA calculated)") parser.add_option("--symmetrize", action="store_true", default=False, help="Prepare input stack for handling symmetry (Default False)") parser.add_option("--overhead", type ="float", default=0.5, help="python overhead per CPU.") (options,args) = parser.parse_args() ##### from mpi import mpi_init, mpi_comm_rank, mpi_comm_size, mpi_recv, MPI_COMM_WORLD from mpi import mpi_barrier, mpi_reduce, mpi_bcast, mpi_send, MPI_FLOAT, MPI_SUM, MPI_INT, MPI_MAX #from mpi import * from applications import MPI_start_end from reconstruction import recons3d_em, recons3d_em_MPI from reconstruction import recons3d_4nn_MPI, recons3d_4nn_ctf_MPI from utilities import print_begin_msg, print_end_msg, print_msg from utilities import read_text_row, get_image, get_im, wrap_mpi_send, wrap_mpi_recv from utilities import bcast_EMData_to_all, bcast_number_to_all from utilities import get_symt # This is code for handling symmetries by the above program. To be incorporated. PAP 01/27/2015 from EMAN2db import db_open_dict # Set up global variables related to bdb cache if global_def.CACHE_DISABLE: from utilities import disable_bdb_cache disable_bdb_cache() # Set up global variables related to ERROR function global_def.BATCH = True # detect if program is running under MPI RUNNING_UNDER_MPI = "OMPI_COMM_WORLD_SIZE" in os.environ if RUNNING_UNDER_MPI: global_def.MPI = True if options.output_dir =="./": current_output_dir = os.path.abspath(options.output_dir) else: current_output_dir = options.output_dir if options.symmetrize : if RUNNING_UNDER_MPI: try: sys.argv = mpi_init(len(sys.argv), sys.argv) try: number_of_proc = mpi_comm_size(MPI_COMM_WORLD) if( number_of_proc > 1 ): ERROR("Cannot use more than one CPU for symmetry preparation","sx3dvariability",1) except: pass except: pass if not os.path.exists(current_output_dir): os.mkdir(current_output_dir) # Input #instack = "Clean_NORM_CTF_start_wparams.hdf" #instack = "bdb:data" from logger import Logger,BaseLogger_Files if os.path.exists(os.path.join(current_output_dir, "log.txt")): os.remove(os.path.join(current_output_dir, "log.txt")) log_main=Logger(BaseLogger_Files()) log_main.prefix = os.path.join(current_output_dir, "./") instack = args[0] sym = options.sym.lower() if( sym == "c1" ): ERROR("There is no need to symmetrize stack for C1 symmetry","sx3dvariability",1) line ="" for a in sys.argv: line +=" "+a log_main.add(line) if(instack[:4] !="bdb:"): #if output_dir =="./": stack = "bdb:data" stack = "bdb:"+current_output_dir+"/data" delete_bdb(stack) junk = cmdexecute("sxcpy.py "+instack+" "+stack) else: stack = instack qt = EMUtil.get_all_attributes(stack,'xform.projection') na = len(qt) ts = get_symt(sym) ks = len(ts) angsa = [None]*na for k in range(ks): #Qfile = "Q%1d"%k #if options.output_dir!="./": Qfile = os.path.join(options.output_dir,"Q%1d"%k) Qfile = os.path.join(current_output_dir, "Q%1d"%k) #delete_bdb("bdb:Q%1d"%k) delete_bdb("bdb:"+Qfile) #junk = cmdexecute("e2bdb.py "+stack+" --makevstack=bdb:Q%1d"%k) junk = cmdexecute("e2bdb.py "+stack+" --makevstack=bdb:"+Qfile) #DB = db_open_dict("bdb:Q%1d"%k) DB = db_open_dict("bdb:"+Qfile) for i in range(na): ut = qt[i]*ts[k] DB.set_attr(i, "xform.projection", ut) #bt = ut.get_params("spider") #angsa[i] = [round(bt["phi"],3)%360.0, round(bt["theta"],3)%360.0, bt["psi"], -bt["tx"], -bt["ty"]] #write_text_row(angsa, 'ptsma%1d.txt'%k) #junk = cmdexecute("e2bdb.py "+stack+" --makevstack=bdb:Q%1d"%k) #junk = cmdexecute("sxheader.py bdb:Q%1d --params=xform.projection --import=ptsma%1d.txt"%(k,k)) DB.close() #if options.output_dir =="./": delete_bdb("bdb:sdata") delete_bdb("bdb:" + current_output_dir + "/"+"sdata") #junk = cmdexecute("e2bdb.py . --makevstack=bdb:sdata --filt=Q") sdata = "bdb:"+current_output_dir+"/"+"sdata" print(sdata) junk = cmdexecute("e2bdb.py " + current_output_dir +" --makevstack="+sdata +" --filt=Q") #junk = cmdexecute("ls EMAN2DB/sdata*") #a = get_im("bdb:sdata") a = get_im(sdata) a.set_attr("variabilitysymmetry",sym) #a.write_image("bdb:sdata") a.write_image(sdata) else: from fundamentals import window2d sys.argv = mpi_init(len(sys.argv), sys.argv) myid = mpi_comm_rank(MPI_COMM_WORLD) number_of_proc = mpi_comm_size(MPI_COMM_WORLD) main_node = 0 shared_comm = mpi_comm_split_type(MPI_COMM_WORLD, MPI_COMM_TYPE_SHARED, 0, MPI_INFO_NULL) myid_on_node = mpi_comm_rank(shared_comm) no_of_processes_per_group = mpi_comm_size(shared_comm) masters_from_groups_vs_everything_else_comm = mpi_comm_split(MPI_COMM_WORLD, main_node == myid_on_node, myid_on_node) color, no_of_groups, balanced_processor_load_on_nodes = get_colors_and_subsets(main_node, MPI_COMM_WORLD, myid, \ shared_comm, myid_on_node, masters_from_groups_vs_everything_else_comm) overhead_loading = options.overhead*number_of_proc #memory_per_node = options.memory_per_node #if memory_per_node == -1.: memory_per_node = 2.*no_of_processes_per_group keepgoing = 1 current_window = options.window current_decimate = options.decimate if len(args) == 1: stack = args[0] else: print(( "usage: " + usage)) print(( "Please run '" + progname + " -h' for detailed options")) return 1 t0 = time() # obsolete flags options.MPI = True #options.nvec = 0 options.radiuspca = -1 options.iter = 40 options.abs = 0.0 options.squ = 0.0 if options.fl > 0.0 and options.aa == 0.0: ERROR("Fall off has to be given for the low-pass filter", "sx3dvariability", 1, myid) #if options.VAR and options.SND: # ERROR("Only one of var and SND can be set!", "sx3dvariability", myid) if options.VAR and (options.ave2D or options.ave3D or options.var2D): ERROR("When VAR is set, the program cannot output ave2D, ave3D or var2D", "sx3dvariability", 1, myid) #if options.SND and (options.ave2D or options.ave3D): # ERROR("When SND is set, the program cannot output ave2D or ave3D", "sx3dvariability", 1, myid) #if options.nvec > 0 : # ERROR("PCA option not implemented", "sx3dvariability", 1, myid) #if options.nvec > 0 and options.ave3D == None: # ERROR("When doing PCA analysis, one must set ave3D", "sx3dvariability", 1, myid) if current_decimate>1.0 or current_decimate<0.0: ERROR("Decimate rate should be a value between 0.0 and 1.0", "sx3dvariability", 1, myid) if current_window < 0.0: ERROR("Target window size should be always larger than zero", "sx3dvariability", 1, myid) if myid == main_node: img = get_image(stack, 0) nx = img.get_xsize() ny = img.get_ysize() if(min(nx, ny) < current_window): keepgoing = 0 keepgoing = bcast_number_to_all(keepgoing, main_node, MPI_COMM_WORLD) if keepgoing == 0: ERROR("The target window size cannot be larger than the size of decimated image", "sx3dvariability", 1, myid) import string options.sym = options.sym.lower() # if global_def.CACHE_DISABLE: # from utilities import disable_bdb_cache # disable_bdb_cache() # global_def.BATCH = True if myid == main_node: if not os.path.exists(current_output_dir): os.mkdir(current_output_dir)# Never delete output_dir in the program! img_per_grp = options.img_per_grp #nvec = options.nvec radiuspca = options.radiuspca from logger import Logger,BaseLogger_Files #if os.path.exists(os.path.join(options.output_dir, "log.txt")): os.remove(os.path.join(options.output_dir, "log.txt")) log_main=Logger(BaseLogger_Files()) log_main.prefix = os.path.join(current_output_dir, "./") if myid == main_node: line = "" for a in sys.argv: line +=" "+a log_main.add(line) log_main.add("-------->>>Settings given by all options<<<-------") log_main.add("Symmetry : %s"%options.sym) log_main.add("Input stack : %s"%stack) log_main.add("Output_dir : %s"%current_output_dir) if options.ave3D: log_main.add("Ave3d : %s"%options.ave3D) if options.var3D: log_main.add("Var3d : %s"%options.var3D) if options.ave2D: log_main.add("Ave2D : %s"%options.ave2D) if options.var2D: log_main.add("Var2D : %s"%options.var2D) if options.VAR: log_main.add("VAR : True") else: log_main.add("VAR : False") if options.CTF: log_main.add("CTF correction : True ") else: log_main.add("CTF correction : False ") log_main.add("Image per group : %5d"%options.img_per_grp) log_main.add("Image decimate rate : %4.3f"%current_decimate) log_main.add("Low pass filter : %4.3f"%options.fl) current_fl = options.fl if current_fl == 0.0: current_fl = 0.5 log_main.add("Current low pass filter is equivalent to cutoff frequency %4.3f for original image size"%round((current_fl*current_decimate),3)) log_main.add("Window size : %5d "%current_window) log_main.add("sx3dvariability begins") symbaselen = 0 if myid == main_node: nima = EMUtil.get_image_count(stack) img = get_image(stack) nx = img.get_xsize() ny = img.get_ysize() nnxo = nx nnyo = ny if options.sym != "c1" : imgdata = get_im(stack) try: i = imgdata.get_attr("variabilitysymmetry").lower() if(i != options.sym): ERROR("The symmetry provided does not agree with the symmetry of the input stack", "sx3dvariability", 1, myid) except: ERROR("Input stack is not prepared for symmetry, please follow instructions", "sx3dvariability", 1, myid) from utilities import get_symt i = len(get_symt(options.sym)) if((nima/i)*i != nima): ERROR("The length of the input stack is incorrect for symmetry processing", "sx3dvariability", 1, myid) symbaselen = nima/i else: symbaselen = nima else: nima = 0 nx = 0 ny = 0 nnxo = 0 nnyo = 0 nima = bcast_number_to_all(nima) nx = bcast_number_to_all(nx) ny = bcast_number_to_all(ny) nnxo = bcast_number_to_all(nnxo) nnyo = bcast_number_to_all(nnyo) if current_window > max(nx, ny): ERROR("Window size is larger than the original image size", "sx3dvariability", 1) if current_decimate == 1.: if current_window !=0: nx = current_window ny = current_window else: if current_window == 0: nx = int(nx*current_decimate+0.5) ny = int(ny*current_decimate+0.5) else: nx = int(current_window*current_decimate+0.5) ny = nx symbaselen = bcast_number_to_all(symbaselen) # check FFT prime number from fundamentals import smallprime is_fft_friendly = (nx == smallprime(nx)) if not is_fft_friendly: if myid == main_node: log_main.add("The target image size is not a product of small prime numbers") log_main.add("Program adjusts the input settings!") ### two cases if current_decimate == 1.: nx = smallprime(nx) ny = nx current_window = nx # update if myid == main_node: log_main.add("The window size is updated to %d."%current_window) else: if current_window == 0: nx = smallprime(int(nx*current_decimate+0.5)) current_decimate = float(nx)/nnxo ny = nx if (myid == main_node): log_main.add("The decimate rate is updated to %f."%current_decimate) else: nx = smallprime(int(current_window*current_decimate+0.5)) ny = nx current_window = int(nx/current_decimate+0.5) if (myid == main_node): log_main.add("The window size is updated to %d."%current_window) if myid == main_node: log_main.add("The target image size is %d"%nx) if radiuspca == -1: radiuspca = nx/2-2 if myid == main_node: log_main.add("%-70s: %d\n"%("Number of projection", nima)) img_begin, img_end = MPI_start_end(nima, number_of_proc, myid) """ if options.SND: from projection import prep_vol, prgs from statistics import im_diff from utilities import get_im, model_circle, get_params_proj, set_params_proj from utilities import get_ctf, generate_ctf from filter import filt_ctf imgdata = EMData.read_images(stack, range(img_begin, img_end)) if options.CTF: vol = recons3d_4nn_ctf_MPI(myid, imgdata, 1.0, symmetry=options.sym, npad=options.npad, xysize=-1, zsize=-1) else: vol = recons3d_4nn_MPI(myid, imgdata, symmetry=options.sym, npad=options.npad, xysize=-1, zsize=-1) bcast_EMData_to_all(vol, myid) volft, kb = prep_vol(vol) mask = model_circle(nx/2-2, nx, ny) varList = [] for i in xrange(img_begin, img_end): phi, theta, psi, s2x, s2y = get_params_proj(imgdata[i-img_begin]) ref_prj = prgs(volft, kb, [phi, theta, psi, -s2x, -s2y]) if options.CTF: ctf_params = get_ctf(imgdata[i-img_begin]) ref_prj = filt_ctf(ref_prj, generate_ctf(ctf_params)) diff, A, B = im_diff(ref_prj, imgdata[i-img_begin], mask) diff2 = diff*diff set_params_proj(diff2, [phi, theta, psi, s2x, s2y]) varList.append(diff2) mpi_barrier(MPI_COMM_WORLD) """ if options.VAR: # 2D variance images have no shifts #varList = EMData.read_images(stack, range(img_begin, img_end)) from EMAN2 import Region for index_of_particle in range(img_begin,img_end): image = get_im(stack, index_of_proj) if current_window > 0: varList.append(fdecimate(window2d(image,current_window,current_window), nx,ny)) else: varList.append(fdecimate(image, nx,ny)) else: from utilities import bcast_number_to_all, bcast_list_to_all, send_EMData, recv_EMData from utilities import set_params_proj, get_params_proj, params_3D_2D, get_params2D, set_params2D, compose_transform2 from utilities import model_blank, nearest_proj, model_circle, write_text_row, wrap_mpi_gatherv from applications import pca from statistics import avgvar, avgvar_ctf, ccc from filter import filt_tanl from morphology import threshold, square_root from projection import project, prep_vol, prgs from sets import Set from utilities import wrap_mpi_recv, wrap_mpi_bcast, wrap_mpi_send import numpy as np if myid == main_node: t1 = time() proj_angles = [] aveList = [] tab = EMUtil.get_all_attributes(stack, 'xform.projection') for i in range(nima): t = tab[i].get_params('spider') phi = t['phi'] theta = t['theta'] psi = t['psi'] x = theta if x > 90.0: x = 180.0 - x x = x*10000+psi proj_angles.append([x, t['phi'], t['theta'], t['psi'], i]) t2 = time() log_main.add( "%-70s: %d\n"%("Number of neighboring projections", img_per_grp)) log_main.add("...... Finding neighboring projections\n") log_main.add( "Number of images per group: %d"%img_per_grp) log_main.add( "Now grouping projections") proj_angles.sort() proj_angles_list = np.full((nima, 4), 0.0, dtype=np.float32) for i in range(nima): proj_angles_list[i][0] = proj_angles[i][1] proj_angles_list[i][1] = proj_angles[i][2] proj_angles_list[i][2] = proj_angles[i][3] proj_angles_list[i][3] = proj_angles[i][4] else: proj_angles_list = 0 proj_angles_list = wrap_mpi_bcast(proj_angles_list, main_node, MPI_COMM_WORLD) proj_angles = [] for i in range(nima): proj_angles.append([proj_angles_list[i][0], proj_angles_list[i][1], proj_angles_list[i][2], int(proj_angles_list[i][3])]) del proj_angles_list proj_list, mirror_list = nearest_proj(proj_angles, img_per_grp, range(img_begin, img_end)) all_proj = Set() for im in proj_list: for jm in im: all_proj.add(proj_angles[jm][3]) all_proj = list(all_proj) index = {} for i in range(len(all_proj)): index[all_proj[i]] = i mpi_barrier(MPI_COMM_WORLD) if myid == main_node: log_main.add("%-70s: %.2f\n"%("Finding neighboring projections lasted [s]", time()-t2)) log_main.add("%-70s: %d\n"%("Number of groups processed on the main node", len(proj_list))) log_main.add("Grouping projections took: %12.1f [m]"%((time()-t2)/60.)) log_main.add("Number of groups on main node: ", len(proj_list)) mpi_barrier(MPI_COMM_WORLD) if myid == main_node: log_main.add("...... Calculating the stack of 2D variances \n") # Memory estimation. There are two memory consumption peaks # peak 1. Compute ave, var; # peak 2. Var volume reconstruction; # proj_params = [0.0]*(nima*5) aveList = [] varList = [] #if nvec > 0: eigList = [[] for i in range(nvec)] dnumber = len(all_proj)# all neighborhood set for assigned to myid pnumber = len(proj_list)*2. + img_per_grp # aveList and varList tnumber = dnumber+pnumber vol_size2 = nx**3*4.*8/1.e9 vol_size1 = 2.*nnxo**3*4.*8/1.e9 proj_size = nnxo*nnyo*len(proj_list)*4.*2./1.e9 # both aveList and varList orig_data_size = nnxo*nnyo*4.*tnumber/1.e9 reduced_data_size = nx*nx*4.*tnumber/1.e9 full_data = np.full((number_of_proc, 2), -1., dtype=np.float16) full_data[myid] = orig_data_size, reduced_data_size if myid != main_node: wrap_mpi_send(full_data, main_node, MPI_COMM_WORLD) if myid == main_node: for iproc in range(number_of_proc): if iproc != main_node: dummy = wrap_mpi_recv(iproc, MPI_COMM_WORLD) full_data[np.where(dummy>-1)] = dummy[np.where(dummy>-1)] del dummy mpi_barrier(MPI_COMM_WORLD) full_data = wrap_mpi_bcast(full_data, main_node, MPI_COMM_WORLD) # find the CPU with heaviest load minindx = np.argsort(full_data, 0) heavy_load_myid = minindx[-1][1] total_mem = sum(full_data) if myid == main_node: if current_window == 0: log_main.add("Nx: current image size = %d. Decimated by %f from %d"%(nx, current_decimate, nnxo)) else: log_main.add("Nx: current image size = %d. Windowed to %d, and decimated by %f from %d"%(nx, current_window, current_decimate, nnxo)) log_main.add("Nproj: number of particle images.") log_main.add("Navg: number of 2D average images.") log_main.add("Nvar: number of 2D variance images.") log_main.add("Img_per_grp: user defined image per group for averaging = %d"%img_per_grp) log_main.add("Overhead: total python overhead memory consumption = %f"%overhead_loading) log_main.add("Total memory) = 4.0*nx^2*(nproj + navg +nvar+ img_per_grp)/1.0e9 + overhead: %12.3f [GB]"%\ (total_mem[1] + overhead_loading)) del full_data mpi_barrier(MPI_COMM_WORLD) if myid == heavy_load_myid: log_main.add("Begin reading and preprocessing images on processor. Wait... ") ttt = time() #imgdata = EMData.read_images(stack, all_proj) imgdata = [ None for im in range(len(all_proj))] for index_of_proj in range(len(all_proj)): #image = get_im(stack, all_proj[index_of_proj]) if( current_window > 0): imgdata[index_of_proj] = fdecimate(window2d(get_im(stack, all_proj[index_of_proj]),current_window,current_window), nx, ny) else: imgdata[index_of_proj] = fdecimate(get_im(stack, all_proj[index_of_proj]), nx, ny) if (current_decimate> 0.0 and options.CTF): ctf = imgdata[index_of_proj].get_attr("ctf") ctf.apix = ctf.apix/current_decimate imgdata[index_of_proj].set_attr("ctf", ctf) if myid == heavy_load_myid and index_of_proj%100 == 0: log_main.add(" ...... %6.2f%% "%(index_of_proj/float(len(all_proj))*100.)) mpi_barrier(MPI_COMM_WORLD) if myid == heavy_load_myid: log_main.add("All_proj preprocessing cost %7.2f m"%((time()-ttt)/60.)) log_main.add("Wait untill reading on all CPUs done...") ''' imgdata2 = EMData.read_images(stack, range(img_begin, img_end)) if options.fl > 0.0: for k in xrange(len(imgdata2)): imgdata2[k] = filt_tanl(imgdata2[k], options.fl, options.aa) if options.CTF: vol = recons3d_4nn_ctf_MPI(myid, imgdata2, 1.0, symmetry=options.sym, npad=options.npad, xysize=-1, zsize=-1) else: vol = recons3d_4nn_MPI(myid, imgdata2, symmetry=options.sym, npad=options.npad, xysize=-1, zsize=-1) if myid == main_node: vol.write_image("vol_ctf.hdf") print_msg("Writing to the disk volume reconstructed from averages as : %s\n"%("vol_ctf.hdf")) del vol, imgdata2 mpi_barrier(MPI_COMM_WORLD) ''' from applications import prepare_2d_forPCA from utilities import model_blank from EMAN2 import Transform if not options.no_norm: mask = model_circle(nx/2-2, nx, nx) if options.CTF: from utilities import pad from filter import filt_ctf from filter import filt_tanl if myid == heavy_load_myid: log_main.add("Start computing 2D aveList and varList. Wait...") ttt = time() inner=nx//2-4 outer=inner+2 xform_proj_for_2D = [ None for i in range(len(proj_list))] for i in range(len(proj_list)): ki = proj_angles[proj_list[i][0]][3] if ki >= symbaselen: continue mi = index[ki] dpar = Util.get_transform_params(imgdata[mi], "xform.projection", "spider") phiM, thetaM, psiM, s2xM, s2yM = dpar["phi"],dpar["theta"],dpar["psi"],-dpar["tx"]*current_decimate,-dpar["ty"]*current_decimate grp_imgdata = [] for j in range(img_per_grp): mj = index[proj_angles[proj_list[i][j]][3]] cpar = Util.get_transform_params(imgdata[mj], "xform.projection", "spider") alpha, sx, sy, mirror = params_3D_2D_NEW(cpar["phi"], cpar["theta"],cpar["psi"], -cpar["tx"]*current_decimate, -cpar["ty"]*current_decimate, mirror_list[i][j]) if thetaM <= 90: if mirror == 0: alpha, sx, sy, scale = compose_transform2(alpha, sx, sy, 1.0, phiM - cpar["phi"], 0.0, 0.0, 1.0) else: alpha, sx, sy, scale = compose_transform2(alpha, sx, sy, 1.0, 180-(phiM - cpar["phi"]), 0.0, 0.0, 1.0) else: if mirror == 0: alpha, sx, sy, scale = compose_transform2(alpha, sx, sy, 1.0, -(phiM- cpar["phi"]), 0.0, 0.0, 1.0) else: alpha, sx, sy, scale = compose_transform2(alpha, sx, sy, 1.0, -(180-(phiM - cpar["phi"])), 0.0, 0.0, 1.0) imgdata[mj].set_attr("xform.align2d", Transform({"type":"2D","alpha":alpha,"tx":sx,"ty":sy,"mirror":mirror,"scale":1.0})) grp_imgdata.append(imgdata[mj]) if not options.no_norm: for k in range(img_per_grp): ave, std, minn, maxx = Util.infomask(grp_imgdata[k], mask, False) grp_imgdata[k] -= ave grp_imgdata[k] /= std if options.fl > 0.0: for k in range(img_per_grp): grp_imgdata[k] = filt_tanl(grp_imgdata[k], options.fl, options.aa) # Because of background issues, only linear option works. if options.CTF: ave, var = aves_wiener(grp_imgdata, SNR = 1.0e5, interpolation_method = "linear") else: ave, var = ave_var(grp_imgdata) # Switch to std dev # threshold is not really needed,it is just in case due to numerical accuracy something turns out negative. var = square_root(threshold(var)) set_params_proj(ave, [phiM, thetaM, 0.0, 0.0, 0.0]) set_params_proj(var, [phiM, thetaM, 0.0, 0.0, 0.0]) aveList.append(ave) varList.append(var) xform_proj_for_2D[i] = [phiM, thetaM, 0.0, 0.0, 0.0] ''' if nvec > 0: eig = pca(input_stacks=grp_imgdata, subavg="", mask_radius=radiuspca, nvec=nvec, incore=True, shuffle=False, genbuf=True) for k in range(nvec): set_params_proj(eig[k], [phiM, thetaM, 0.0, 0.0, 0.0]) eigList[k].append(eig[k]) """ if myid == 0 and i == 0: for k in xrange(nvec): eig[k].write_image("eig.hdf", k) """ ''' if (myid == heavy_load_myid) and (i%100 == 0): log_main.add(" ......%6.2f%% "%(i/float(len(proj_list))*100.)) del imgdata, grp_imgdata, cpar, dpar, all_proj, proj_angles, index if not options.no_norm: del mask if myid == main_node: del tab # At this point, all averages and variances are computed mpi_barrier(MPI_COMM_WORLD) if (myid == heavy_load_myid): log_main.add("Computing aveList and varList took %12.1f [m]"%((time()-ttt)/60.)) xform_proj_for_2D = wrap_mpi_gatherv(xform_proj_for_2D, main_node, MPI_COMM_WORLD) if (myid == main_node): write_text_row(xform_proj_for_2D, os.path.join(current_output_dir, "params.txt")) del xform_proj_for_2D mpi_barrier(MPI_COMM_WORLD) if options.ave2D: from fundamentals import fpol from applications import header if myid == main_node: log_main.add("Compute ave2D ... ") km = 0 for i in range(number_of_proc): if i == main_node : for im in range(len(aveList)): aveList[im].write_image(os.path.join(current_output_dir, options.ave2D), km) km += 1 else: nl = mpi_recv(1, MPI_INT, i, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) nl = int(nl[0]) for im in range(nl): ave = recv_EMData(i, im+i+70000) """ nm = mpi_recv(1, MPI_INT, i, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) nm = int(nm[0]) members = mpi_recv(nm, MPI_INT, i, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) ave.set_attr('members', map(int, members)) members = mpi_recv(nm, MPI_FLOAT, i, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) ave.set_attr('pix_err', map(float, members)) members = mpi_recv(3, MPI_FLOAT, i, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) ave.set_attr('refprojdir', map(float, members)) """ tmpvol=fpol(ave, nx, nx,1) tmpvol.write_image(os.path.join(current_output_dir, options.ave2D), km) km += 1 else: mpi_send(len(aveList), 1, MPI_INT, main_node, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) for im in range(len(aveList)): send_EMData(aveList[im], main_node,im+myid+70000) """ members = aveList[im].get_attr('members') mpi_send(len(members), 1, MPI_INT, main_node, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) mpi_send(members, len(members), MPI_INT, main_node, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) members = aveList[im].get_attr('pix_err') mpi_send(members, len(members), MPI_FLOAT, main_node, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) try: members = aveList[im].get_attr('refprojdir') mpi_send(members, 3, MPI_FLOAT, main_node, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) except: mpi_send([-999.0,-999.0,-999.0], 3, MPI_FLOAT, main_node, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) """ if myid == main_node: header(os.path.join(current_output_dir, options.ave2D), params='xform.projection', fimport = os.path.join(current_output_dir, "params.txt")) mpi_barrier(MPI_COMM_WORLD) if options.ave3D: from fundamentals import fpol t5 = time() if myid == main_node: log_main.add("Reconstruct ave3D ... ") ave3D = recons3d_4nn_MPI(myid, aveList, symmetry=options.sym, npad=options.npad) bcast_EMData_to_all(ave3D, myid) if myid == main_node: if current_decimate != 1.0: ave3D = resample(ave3D, 1./current_decimate) ave3D = fpol(ave3D, nnxo, nnxo, nnxo) # always to the orignal image size set_pixel_size(ave3D, 1.0) ave3D.write_image(os.path.join(current_output_dir, options.ave3D)) log_main.add("Ave3D reconstruction took %12.1f [m]"%((time()-t5)/60.0)) log_main.add("%-70s: %s\n"%("The reconstructed ave3D is saved as ", options.ave3D)) mpi_barrier(MPI_COMM_WORLD) del ave, var, proj_list, stack, alpha, sx, sy, mirror, aveList ''' if nvec > 0: for k in range(nvec): if myid == main_node:log_main.add("Reconstruction eigenvolumes", k) cont = True ITER = 0 mask2d = model_circle(radiuspca, nx, nx) while cont: #print "On node %d, iteration %d"%(myid, ITER) eig3D = recons3d_4nn_MPI(myid, eigList[k], symmetry=options.sym, npad=options.npad) bcast_EMData_to_all(eig3D, myid, main_node) if options.fl > 0.0: eig3D = filt_tanl(eig3D, options.fl, options.aa) if myid == main_node: eig3D.write_image(os.path.join(options.outpout_dir, "eig3d_%03d.hdf"%(k, ITER))) Util.mul_img( eig3D, model_circle(radiuspca, nx, nx, nx) ) eig3Df, kb = prep_vol(eig3D) del eig3D cont = False icont = 0 for l in range(len(eigList[k])): phi, theta, psi, s2x, s2y = get_params_proj(eigList[k][l]) proj = prgs(eig3Df, kb, [phi, theta, psi, s2x, s2y]) cl = ccc(proj, eigList[k][l], mask2d) if cl < 0.0: icont += 1 cont = True eigList[k][l] *= -1.0 u = int(cont) u = mpi_reduce([u], 1, MPI_INT, MPI_MAX, main_node, MPI_COMM_WORLD) icont = mpi_reduce([icont], 1, MPI_INT, MPI_SUM, main_node, MPI_COMM_WORLD) if myid == main_node: u = int(u[0]) log_main.add(" Eigenvector: ",k," number changed ",int(icont[0])) else: u = 0 u = bcast_number_to_all(u, main_node) cont = bool(u) ITER += 1 del eig3Df, kb mpi_barrier(MPI_COMM_WORLD) del eigList, mask2d ''' if options.ave3D: del ave3D if options.var2D: from fundamentals import fpol from applications import header if myid == main_node: log_main.add("Compute var2D...") km = 0 for i in range(number_of_proc): if i == main_node : for im in range(len(varList)): tmpvol=fpol(varList[im], nx, nx,1) tmpvol.write_image(os.path.join(current_output_dir, options.var2D), km) km += 1 else: nl = mpi_recv(1, MPI_INT, i, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) nl = int(nl[0]) for im in range(nl): ave = recv_EMData(i, im+i+70000) tmpvol=fpol(ave, nx, nx,1) tmpvol.write_image(os.path.join(current_output_dir, options.var2D), km) km += 1 else: mpi_send(len(varList), 1, MPI_INT, main_node, SPARX_MPI_TAG_UNIVERSAL, MPI_COMM_WORLD) for im in range(len(varList)): send_EMData(varList[im], main_node, im+myid+70000)# What with the attributes?? mpi_barrier(MPI_COMM_WORLD) if myid == main_node: from applications import header header(os.path.join(current_output_dir, options.var2D), params = 'xform.projection',fimport = os.path.join(current_output_dir, "params.txt")) mpi_barrier(MPI_COMM_WORLD) if options.var3D: if myid == main_node: log_main.add("Reconstruct var3D ...") t6 = time() # radiusvar = options.radius # if( radiusvar < 0 ): radiusvar = nx//2 -3 res = recons3d_4nn_MPI(myid, varList, symmetry = options.sym, npad=options.npad) #res = recons3d_em_MPI(varList, vol_stack, options.iter, radiusvar, options.abs, True, options.sym, options.squ) if myid == main_node: from fundamentals import fpol if current_decimate != 1.0: res = resample(res, 1./current_decimate) res = fpol(res, nnxo, nnxo, nnxo) set_pixel_size(res, 1.0) res.write_image(os.path.join(current_output_dir, options.var3D)) log_main.add("%-70s: %s\n"%("The reconstructed var3D is saved as ", options.var3D)) log_main.add("Var3D reconstruction took %f12.1 [m]"%((time()-t6)/60.0)) log_main.add("Total computation time %f12.1 [m]"%((time()-t0)/60.0)) log_main.add("sx3dvariability finishes") from mpi import mpi_finalize mpi_finalize() if RUNNING_UNDER_MPI: global_def.MPI = False global_def.BATCH = False
def found_outliers(list_of_projection_indices, outlier_percentile, rviper_iter, masterdir, bdb_stack_location, outlier_index_threshold_method, angle_threshold): # sxheader.py bdb:nj --consecutive --params=OID import numpy as np mainoutputdir = masterdir + DIR_DELIM + NAME_OF_MAIN_DIR + ("%03d" + DIR_DELIM) %(rviper_iter) # if this data analysis step was already performed in the past then return for check_run in list_of_projection_indices: if not (os.path.exists(mainoutputdir + DIR_DELIM + NAME_OF_RUN_DIR + "%03d"%(check_run) + DIR_DELIM + "rotated_reduced_params.txt")): break else: return print "identify_outliers" projs = [] for i1 in list_of_projection_indices: projs.append(read_text_row(mainoutputdir + NAME_OF_RUN_DIR + "%03d"%(i1) + DIR_DELIM + "params.txt")) # ti1, ti3, out = find_common_subset(projs, 1.0) subset, avg_diff_per_image, rotated_params = find_common_subset(projs, target_threshold = 0) # subset, avg_diff_per_image, rotated_params = find_common_subset(projs, target_threshold = 1.0) error_values_and_indices = [] for i in xrange(len(avg_diff_per_image)): error_values_and_indices.append([avg_diff_per_image[i], i]) del subset, avg_diff_per_image error_values_and_indices.sort() if outlier_index_threshold_method == "discontinuity_in_derivative": outlier_index_threshold = find_index_of_discontinuity_in_derivative([i[0] for i in error_values_and_indices], list_of_projection_indices, mainoutputdir, outlier_percentile) elif outlier_index_threshold_method == "percentile": outlier_index_threshold = outlier_percentile * (len(error_values_and_indices) - 1)/ 100.0 elif outlier_index_threshold_method == "angle_measure": error_values = [i[0] for i in error_values_and_indices] outlier_index_threshold = min(range(len(error_values)), key=lambda i: abs(error_values[i]-angle_threshold)) elif outlier_index_threshold_method == "use all images": outlier_index_threshold = len(error_values_and_indices) index_keep_images = [i[1] for i in error_values_and_indices[:outlier_index_threshold]] index_outliers = [i[1] for i in error_values_and_indices[outlier_index_threshold:]] # print "error_values_and_indices: %f"%error_values_and_indices print "index_outliers: ", index_outliers import copy reversed_sorted_index_outliers = copy.deepcopy(index_outliers) reversed_sorted_index_outliers.sort(reverse=True) for k in xrange(len(projs)): for l in reversed_sorted_index_outliers: del rotated_params[k][l] index_outliers.sort() index_keep_images.sort() write_text_file(index_outliers, mainoutputdir + "this_iteration_index_outliers.txt") write_text_file(index_keep_images, mainoutputdir + "this_iteration_index_keep_images.txt") #if len(index_outliers) < 3: #return False if len(index_outliers) > 0: cmd = "{} {} {} {}".format("e2bdb.py ", bdb_stack_location + "_%03d"%(rviper_iter - 1), "--makevstack=" + bdb_stack_location + "_outliers_%03d"%(rviper_iter), "--list=" + mainoutputdir + "this_iteration_index_outliers.txt") cmdexecute(cmd) cmd = "{} {} {} {}".format("e2bdb.py ", bdb_stack_location + "_%03d"%(rviper_iter - 1), "--makevstack=" + bdb_stack_location + "_%03d"%(rviper_iter), "--list=" + mainoutputdir + "this_iteration_index_keep_images.txt") cmdexecute(cmd) dat = EMData.read_images(bdb_stack_location + "_%03d"%(rviper_iter - 1)) write_text_file([dat[i].get_attr("original_image_index") for i in index_outliers],mainoutputdir + "index_outliers.txt") write_text_file([dat[i].get_attr("original_image_index") for i in index_keep_images],mainoutputdir + "index_keep_images.txt") print "index_outliers:: " + str(index_outliers) # write rotated param files for i1 in range(len(list_of_projection_indices)): write_text_row(rotated_params[i1], mainoutputdir + NAME_OF_RUN_DIR + "%03d"%(list_of_projection_indices[i1]) + DIR_DELIM + "rotated_reduced_params.txt") return True
def main(): import os import sys from optparse import OptionParser from global_def import SPARXVERSION import global_def arglist = [] for arg in sys.argv: arglist.append(arg) progname = os.path.basename(arglist[0]) usage2 = progname + """ inputfile outputfile [options] Functionalities: 1. Helicise input volume and save the result to output volume: sxhelicon_utils.py input_vol.hdf output_vol.hdf --helicise --dp=27.6 --dphi=166.5 --fract=0.65 --rmax=70 --rmin=1 --apix=1.84 --sym=D1 2. Helicise pdb file and save the result to a new pdb file: sxhelicon_utils.py input.pdb output.pdb --helicisepdb --dp=27.6 --dphi=166.5 --nrepeats --apix=1.84 3. Generate two lists of image indices used to split segment stack into halves for helical fsc calculation. sxhelicon_utils.py bdb:big_stack --hfsc='flst' --filament_attr=filament 4. Map of filament distribution in the stack sxhelicon_utils.py bdb:big_stack --filinfo=info.txt The output file will contain four columns: 1 2 3 4 first image number last image number number of images in the filament name 5. Predict segments' orientation parameters based on distances between segments and known helical symmetry sxhelicon_utils.py bdb:big_stack --predict_helical=helical_params.txt --dp=27.6 --dphi=166.5 --apix=1.84 6. Generate disks from filament based reconstructions: sxheader.py stk.hdf --params=xform.projection --import=params.txt mpirun -np 2 sxhelicon_utils.py stk.hdf --gendisk='bdb:disk' --ref_nx=100 --ref_ny=100 --ref_nz=200 --apix=1.84 --dp=27.6 --dphi=166.715 --fract=0.67 --rmin=0 --rmax=64 --function="[.,nofunc,helical3c]" --sym="c1" --MPI 7. Stack disks based on helical symmetry parameters sxhelicon_utils.py disk_to_stack.hdf --stackdisk=stacked_disks.hdf --dphi=166.5 --dp=27.6 --ref_nx=160 --ref_ny=160 --ref_nz=225 --apix=1.84 8. Helical symmetry search: mpirun -np 3 sxhelicon_utils.py volf0010.hdf outsymsearch --symsearch --dp=27.6 --dphi=166.715 --apix=1.84 --fract=0.65 --rmin=0 --rmax=92.0 --datasym=datasym.txt --dp_step=0.92 --ndp=3 --dphi_step=1.0 --ndphi=10 --MPI """ parser = OptionParser(usage2, version=SPARXVERSION) #parser.add_option("--ir", type="float", default= -1, help="inner radius for rotational correlation > 0 (set to 1) (Angstroms)") parser.add_option( "--ou", type="float", default=-1, help= "outer radius for rotational 2D correlation < int(nx/2)-1 (set to the radius of the particle) (Angstroms)" ) parser.add_option( "--rs", type="int", default=1, help="step between rings in rotational correlation >0 (set to 1)") parser.add_option( "--xr", type="string", default="4 2 1 1 1", help= "range for translation search in x direction, search is +/-xr (Angstroms) " ) parser.add_option( "--txs", type="string", default="1 1 1 0.5 0.25", help= "step size of the translation search in x directions, search is -xr, -xr+ts, 0, xr-ts, xr (Angstroms)" ) parser.add_option("--delta", type="string", default="10 6 4 3 2", help="angular step of reference projections") parser.add_option("--an", type="string", default="-1", help="angular neighborhood for local searches") parser.add_option( "--maxit", type="int", default=30, help= "maximum number of iterations performed for each angular step (set to 30) " ) parser.add_option("--CTF", action="store_true", default=False, help="CTF correction") parser.add_option("--snr", type="float", default=1.0, help="Signal-to-Noise Ratio of the data") parser.add_option("--MPI", action="store_true", default=False, help="use MPI version") #parser.add_option("--fourvar", action="store_true", default=False, help="compute Fourier variance") parser.add_option("--apix", type="float", default=-1.0, help="pixel size in Angstroms") parser.add_option("--dp", type="float", default=-1.0, help="delta z - translation in Angstroms") parser.add_option("--dphi", type="float", default=-1.0, help="delta phi - rotation in degrees") parser.add_option("--rmin", type="float", default=0.0, help="minimal radius for hsearch (Angstroms)") parser.add_option("--rmax", type="float", default=80.0, help="maximal radius for hsearch (Angstroms)") parser.add_option("--fract", type="float", default=0.7, help="fraction of the volume used for helical search") parser.add_option("--sym", type="string", default="c1", help="symmetry of the structure") parser.add_option("--function", type="string", default="helical", help="name of the reference preparation function") parser.add_option("--npad", type="int", default=2, help="padding size for 3D reconstruction") parser.add_option("--debug", action="store_true", default=False, help="debug") parser.add_option("--volalixshift", action="store_true", default=False, help="Use volalixshift refinement") parser.add_option( "--searchxshift", type="float", default=0.0, help= "search range for x-shift determination: +/- searchxshift (Angstroms)") parser.add_option( "--nearby", type="float", default=6.0, help= "neighborhood within which to search for peaks in 1D ccf for x-shift search (Angstroms)" ) # filinfo parser.add_option( "--filinfo", type="string", default="", help= "Store in an output text file infomration about distribution of filaments in the stack." ) # diskali parser.add_option("--diskali", action="store_true", default=False, help="volume alignment") parser.add_option( "--zstep", type="float", default=1, help="Step size for translational search along z (Angstroms)") # helicise parser.add_option( "--helicise", action="store_true", default=False, help="helicise input volume and save results to output volume") parser.add_option( "--hfsc", type="string", default="", help= "Generate two lists of image indices used to split segment stack into halves for helical fsc calculation. The lists will be stored in two text files named using file_prefix with '_even' and '_odd' suffixes, respectively." ) parser.add_option( "--filament_attr", type="string", default="filament", help="attribute under which filament identification is stored") parser.add_option( "--predict_helical", type="string", default="", help="Generate projection parameters consistent with helical symmetry") # helicise pdb parser.add_option( "--helicisepdb", action="store_true", default=False, help="Helicise pdb file and save the result to a new pdb file") parser.add_option( "--nrepeats", type="int", default=50, help= "Number of time the helical symmetry will be applied to the input file" ) # input options for generating disks parser.add_option( "--gendisk", type="string", default="", help="Name of file under which generated disks will be saved to") parser.add_option("--ref_nx", type="int", default=-1, help="nx=ny volume size") parser.add_option( "--ref_nz", type="int", default=-1, help="nz volume size - computed disks will be nx x ny x rise/apix") parser.add_option( "--new_pixel_size", type="float", default=-1, help= "desired pixel size of the output disks. The default is -1, in which case there is no resampling (unless --match_pixel_rise flag is True)." ) parser.add_option( "--maxerror", type="float", default=0.1, help= "proportional to the maximum amount of error to tolerate between (dp/new_pixel_size) and int(dp/new_pixel_size ), where new_pixel_size is the pixel size calculated when the option --match_pixel_rise flag is True." ) parser.add_option( "--match_pixel_rise", action="store_true", default=False, help= "calculate new pixel size such that the rise is approximately integer number of pixels given the new pixel size. This will be the pixel size of the output disks." ) # get consistency parser.add_option( "--consistency", type="string", default="", help="Name of parameters to get consistency statistics for") parser.add_option("--phithr", type="float", default=2.0, help="phi threshold for consistency check") parser.add_option("--ythr", type="float", default=2.0, help="y threshold (in Angstroms) for consistency check") parser.add_option( "--segthr", type="int", default=3, help="minimum number of segments/filament for consistency check") # stack disks parser.add_option( "--stackdisk", type="string", default="", help="Name of file under which output volume will be saved to.") parser.add_option("--ref_ny", type="int", default=-1, help="ny of output volume size. Default is ref_nx") # symmetry search parser.add_option("--symsearch", action="store_true", default=False, help="Do helical symmetry search.") parser.add_option( "--ndp", type="int", default=12, help= "In symmetrization search, number of delta z steps equals to 2*ndp+1") parser.add_option( "--ndphi", type="int", default=12, help= "In symmetrization search, number of dphi steps equals to 2*ndphi+1") parser.add_option( "--dp_step", type="float", default=0.1, help="delta z step for symmetrization [Angstroms] (default 0.1)") parser.add_option( "--dphi_step", type="float", default=0.1, help="dphi step for symmetrization [degrees] (default 0.1)") parser.add_option("--datasym", type="string", default="datasym.txt", help="symdoc") parser.add_option( "--symdoc", type="string", default="", help="text file containing helical symmetry parameters dp and dphi") # filament statistics in the stack (options, args) = parser.parse_args(arglist[1:]) if len(args) < 1 or len(args) > 5: print("Various helical reconstruction related functionalities: " + usage2) print("Please run '" + progname + " -h' for detailed options") else: if len(options.hfsc) > 0: if len(args) != 1: print("Incorrect number of parameters") sys.exit() from applications import imgstat_hfsc imgstat_hfsc(args[0], options.hfsc, options.filament_attr) sys.exit() elif len(options.filinfo) > 0: if len(args) != 1: print("Incorrect number of parameters") sys.exit() from EMAN2 import EMUtil filams = EMUtil.get_all_attributes(args[0], "filament") ibeg = 0 filcur = filams[0] n = len(filams) inf = [] i = 1 while (i <= n): if (i < n): fis = filams[i] else: fis = "" if (fis != filcur): iend = i - 1 inf.append([ibeg, iend, iend - ibeg + 1, filcur]) ibeg = i filcur = fis i += 1 from utilities import write_text_row write_text_row(inf, options.filinfo) sys.exit() if len(options.stackdisk) > 0: if len(args) != 1: print("Incorrect number of parameters") sys.exit() dpp = (float(options.dp) / options.apix) rise = int(dpp) if (abs(float(rise) - dpp) > 1.0e-3): print(" dpp has to be integer multiplicity of the pixel size") sys.exit() from utilities import get_im v = get_im(args[0]) from applications import stack_disks ref_ny = options.ref_ny if ref_ny < 0: ref_ny = options.ref_nx sv = stack_disks(v, options.ref_nx, ref_ny, options.ref_nz, options.dphi, rise) sv.write_image(options.stackdisk) sys.exit() if len(options.consistency) > 0: if len(args) != 1: print("Incorrect number of parameters") sys.exit() from development import consistency_params consistency_params(args[0], options.consistency, options.dphi, options.dp, options.apix, phithr=options.phithr, ythr=options.ythr, THR=options.segthr) sys.exit() rminp = int((float(options.rmin) / options.apix) + 0.5) rmaxp = int((float(options.rmax) / options.apix) + 0.5) from utilities import get_input_from_string, get_im xr = get_input_from_string(options.xr) txs = get_input_from_string(options.txs) irp = 1 if options.ou < 0: oup = -1 else: oup = int((options.ou / options.apix) + 0.5) xrp = '' txsp = '' for i in xrange(len(xr)): xrp += " " + str(float(xr[i]) / options.apix) for i in xrange(len(txs)): txsp += " " + str(float(txs[i]) / options.apix) searchxshiftp = int((options.searchxshift / options.apix) + 0.5) nearbyp = int((options.nearby / options.apix) + 0.5) zstepp = int((options.zstep / options.apix) + 0.5) if options.MPI: from mpi import mpi_init, mpi_finalize sys.argv = mpi_init(len(sys.argv), sys.argv) if len(options.predict_helical) > 0: if len(args) != 1: print("Incorrect number of parameters") sys.exit() if options.dp < 0: print( "Helical symmetry paramter rise --dp should not be negative" ) sys.exit() from applications import predict_helical_params predict_helical_params(args[0], options.dp, options.dphi, options.apix, options.predict_helical) sys.exit() if options.helicise: if len(args) != 2: print("Incorrect number of parameters") sys.exit() if options.dp < 0: print( "Helical symmetry paramter rise --dp should not be negative" ) sys.exit() from utilities import get_im, sym_vol vol = get_im(args[0]) vol = sym_vol(vol, options.sym) hvol = vol.helicise(options.apix, options.dp, options.dphi, options.fract, rmaxp, rminp) hvol = sym_vol(hvol, options.sym) hvol.write_image(args[1]) sys.exit() if options.helicisepdb: if len(args) != 2: print("Incorrect number of parameters") sys.exit() if options.dp < 0: print( "Helical symmetry paramter rise --dp should not be negative" ) sys.exit() from math import cos, sin, radians from copy import deepcopy import numpy from numpy import zeros, dot, float32 dp = options.dp dphi = options.dphi nperiod = options.nrepeats infile = open(args[0], "r") pall = infile.readlines() infile.close() p = [] pos = [] lkl = -1 for i in xrange(len(pall)): if ((pall[i])[:4] == 'ATOM'): if (lkl == -1): lkl = i p.append(pall[i]) pos.append(i) n = len(p) X = zeros((3, len(p)), dtype=float32) X_new = zeros((3, len(p)), dtype=float32) for i in xrange(len(p)): element = deepcopy(p[i]) X[0, i] = float(element[30:38]) X[1, i] = float(element[38:46]) X[2, i] = float(element[46:54]) pnew = [] for j in xrange(-nperiod, nperiod + 1): for i in xrange(n): pnew.append(deepcopy(p[i])) dphi = radians(dphi) m = zeros((3, 3), dtype=float32) t = zeros((3, 1), dtype=float32) m[2][2] = 1.0 t[0, 0] = 0.0 t[1, 0] = 0.0 for j in xrange(-nperiod, nperiod + 1): if j != 0: rd = j * dphi m[0][0] = cos(rd) m[0][1] = sin(rd) m[1][0] = -m[0][1] m[1][1] = m[0][0] t[2, 0] = j * dp X_new = dot(m, X) + t for i in xrange(n): pnew[j * n + i] = pnew[j * n + i][:30] + "%8.3f" % (float( X_new[0, i])) + "%8.3f" % (float( X_new[1, i])) + "%8.3f" % (float( X_new[2, i])) + pnew[j * n + i][54:] outfile = open(args[1], "w") outfile.writelines(pall[0:lkl]) outfile.writelines(pnew) outfile.writelines("END\n") outfile.close() sys.exit() if options.volalixshift: if options.maxit > 1: print( "Inner iteration for x-shift determinatin is restricted to 1" ) sys.exit() if len(args) < 4: mask = None else: mask = args[3] from applications import volalixshift_MPI global_def.BATCH = True volalixshift_MPI(args[0], args[1], args[2], searchxshiftp, options.apix, options.dp, options.dphi, options.fract, rmaxp, rminp, mask, options.maxit, options.CTF, options.snr, options.sym, options.function, options.npad, options.debug, nearbyp) global_def.BATCH = False if options.diskali: #if options.maxit > 1: # print "Inner iteration for disk alignment is restricted to 1" # sys.exit() if len(args) < 4: mask = None else: mask = args[3] global_def.BATCH = True if (options.sym[:1] == "d" or options.sym[:1] == "D"): from development import diskaliD_MPI diskaliD_MPI(args[0], args[1], args[2], mask, options.dp, options.dphi, options.apix, options.function, zstepp, options.fract, rmaxp, rminp, options.CTF, options.maxit, options.sym) else: from applications import diskali_MPI diskali_MPI(args[0], args[1], args[2], mask, options.dp, options.dphi, options.apix, options.function, zstepp, options.fract, rmaxp, rminp, options.CTF, options.maxit, options.sym) global_def.BATCH = False if options.symsearch: if len(options.symdoc) < 1: if options.dp < 0 or options.dphi < 0: print( "Enter helical symmetry parameters either using --symdoc or --dp and --dphi" ) sys.exit() if options.dp < 0 or options.dphi < 0: # read helical symmetry parameters from symdoc from utilities import read_text_row hparams = read_text_row(options.symdoc) dp = hparams[0][0] dphi = hparams[0][1] else: dp = options.dp dphi = options.dphi from applications import symsearch_MPI if len(args) < 3: mask = None else: mask = args[2] global_def.BATCH = True symsearch_MPI(args[0], args[1], mask, dp, options.ndp, options.dp_step, dphi, options.ndphi, options.dphi_step, rminp, rmaxp, options.fract, options.sym, options.function, options.datasym, options.apix, options.debug) global_def.BATCH = False elif len(options.gendisk) > 0: from applications import gendisks_MPI global_def.BATCH = True if len(args) == 1: mask3d = None else: mask3d = args[1] if options.dp < 0: print( "Helical symmetry paramter rise --dp must be explictly set!" ) sys.exit() gendisks_MPI(args[0], mask3d, options.ref_nx, options.apix, options.dp, options.dphi, options.fract, rmaxp, rminp, options.CTF, options.function, options.sym, options.gendisk, options.maxerror, options.new_pixel_size, options.match_pixel_rise) global_def.BATCH = False if options.MPI: from mpi import mpi_finalize mpi_finalize()
def main(): from utilities import get_input_from_string progname = os.path.basename(sys.argv[0]) usage = ( progname + " stack output_average --radius=particle_radius --xr=xr --yr=yr --ts=ts --thld_err=thld_err --num_ali=num_ali --fl=fl --aa=aa --CTF --verbose --stables" ) parser = OptionParser(usage, version=SPARXVERSION) parser.add_option("--radius", type="int", default=-1, help=" particle radius for alignment") parser.add_option( "--xr", type="string", default="2 1", help="range for translation search in x direction, search is +/xr (default 2,1)", ) parser.add_option( "--yr", type="string", default="-1", help="range for translation search in y direction, search is +/yr (default = same as xr)", ) parser.add_option( "--ts", type="string", default="1 0.5", help="step size of the translation search in both directions, search is -xr, -xr+ts, 0, xr-ts, xr, can be fractional (default: 1,0.5)", ) parser.add_option("--thld_err", type="float", default=0.75, help="threshld of pixel error (default = 0.75)") parser.add_option( "--num_ali", type="int", default=5, help="number of alignments performed for stability (default = 5)" ) parser.add_option("--maxit", type="int", default=30, help="number of iterations for each xr (default = 30)") parser.add_option( "--fl", type="float", default=0.3, help="cut-off frequency of hyperbolic tangent low-pass Fourier filter (default = 0.3)", ) parser.add_option( "--aa", type="float", default=0.2, help="fall-off of hyperbolic tangent low-pass Fourier filter (default = 0.2)" ) parser.add_option("--CTF", action="store_true", default=False, help="Use CTF correction during the alignment ") parser.add_option( "--verbose", action="store_true", default=False, help="print individual pixel error (default = False)" ) parser.add_option( "--stables", action="store_true", default=False, help="output the stable particles number in file (default = False)", ) parser.add_option( "--method", type="string", default=" ", help="SHC (standard method is default when flag is ommitted)" ) (options, args) = parser.parse_args() if len(args) != 1 and len(args) != 2: print "usage: " + usage print "Please run '" + progname + " -h' for detailed options" else: if global_def.CACHE_DISABLE: from utilities import disable_bdb_cache disable_bdb_cache() from applications import within_group_refinement, ali2d_ras from pixel_error import multi_align_stability from utilities import write_text_file, write_text_row global_def.BATCH = True xrng = get_input_from_string(options.xr) if options.yr == "-1": yrng = xrng else: yrng = get_input_from_string(options.yr) step = get_input_from_string(options.ts) class_data = EMData.read_images(args[0]) nx = class_data[0].get_xsize() ou = options.radius num_ali = options.num_ali if ou == -1: ou = nx / 2 - 2 from utilities import model_circle, get_params2D, set_params2D mask = model_circle(ou, nx, nx) if options.CTF: from filter import filt_ctf for im in xrange(len(class_data)): # Flip phases class_data[im] = filt_ctf(class_data[im], class_data[im].get_attr("ctf"), binary=1) for im in class_data: im.set_attr("previousmax", -1.0e10) try: t = im.get_attr("xform.align2d") # if they are there, no need to set them! except: try: t = im.get_attr("xform.projection") d = t.get_params("spider") set_params2D(im, [0.0, -d["tx"], -d["ty"], 0, 1.0]) except: set_params2D(im, [0.0, 0.0, 0.0, 0, 1.0]) all_ali_params = [] for ii in xrange(num_ali): ali_params = [] if options.verbose: ALPHA = [] SX = [] SY = [] MIRROR = [] if xrng[0] == 0.0 and yrng[0] == 0.0: avet = ali2d_ras( class_data, randomize=True, ir=1, ou=ou, rs=1, step=1.0, dst=90.0, maxit=options.maxit, check_mirror=True, FH=options.fl, FF=options.aa, ) else: avet = within_group_refinement( class_data, mask, True, 1, ou, 1, xrng, yrng, step, 90.0, maxit=options.maxit, FH=options.fl, FF=options.aa, method=options.method, ) from utilities import info # print " avet ",info(avet) for im in class_data: alpha, sx, sy, mirror, scale = get_params2D(im) ali_params.extend([alpha, sx, sy, mirror]) if options.verbose: ALPHA.append(alpha) SX.append(sx) SY.append(sy) MIRROR.append(mirror) all_ali_params.append(ali_params) if options.verbose: write_text_file([ALPHA, SX, SY, MIRROR], "ali_params_run_%d" % ii) """ avet = class_data[0] from utilities import read_text_file all_ali_params = [] for ii in xrange(5): temp = read_text_file( "ali_params_run_%d"%ii,-1) uuu = [] for k in xrange(len(temp[0])): uuu.extend([temp[0][k],temp[1][k],temp[2][k],temp[3][k]]) all_ali_params.append(uuu) """ stable_set, mir_stab_rate, pix_err = multi_align_stability( all_ali_params, 0.0, 10000.0, options.thld_err, options.verbose, 2 * ou + 1 ) print "%4s %20s %20s %20s %30s %6.2f" % ( "", "Size of set", "Size of stable set", "Mirror stab rate", "Pixel error prior to pruning the set above threshold of", options.thld_err, ) print "Average stat: %10d %20d %20.2f %15.2f" % (len(class_data), len(stable_set), mir_stab_rate, pix_err) if len(stable_set) > 0: if options.stables: stab_mem = [[0, 0.0, 0] for j in xrange(len(stable_set))] for j in xrange(len(stable_set)): stab_mem[j] = [int(stable_set[j][1]), stable_set[j][0], j] write_text_row(stab_mem, "stable_particles.txt") stable_set_id = [] particle_pixerr = [] for s in stable_set: stable_set_id.append(s[1]) particle_pixerr.append(s[0]) from fundamentals import rot_shift2D avet.to_zero() l = -1 print "average parameters: angle, x-shift, y-shift, mirror" for j in stable_set_id: l += 1 print " %4d %4d %12.2f %12.2f %12.2f %1d" % ( l, j, stable_set[l][2][0], stable_set[l][2][1], stable_set[l][2][2], int(stable_set[l][2][3]), ) avet += rot_shift2D( class_data[j], stable_set[l][2][0], stable_set[l][2][1], stable_set[l][2][2], stable_set[l][2][3] ) avet /= l + 1 avet.set_attr("members", stable_set_id) avet.set_attr("pix_err", pix_err) avet.set_attr("pixerr", particle_pixerr) avet.write_image(args[1]) global_def.BATCH = False