import os, sys
import subprocess as sub
selfbin = os.path.realpath(os.path.dirname(sys.argv[0]))
sys.path.insert(0, '/home/grigoryanlab/home/fzheng/modules_py')
import General
if len(sys.argv) - 1 != 1:
print '<usage> [a list file]'
exit(0)
lst = sys.argv[1]
for l in open(lst):
info = l.strip().split('/')
name, pds = info[-2], info[-1]
cmap = General.changeExt(pds, 'cmap')
if os.path.isfile(name + '/' + cmap):
sub.call(['python', selfbin + '/threeBodyContactPotential.py', name + '/' + cmap, '0.01'])
par.add_argument('--conR', action = 'store_true', help = 'whether this is for a pair of contact')
par.add_argument('--env', help = 'if not None, also modify an .env file')
args = par.parse_args()
dirs = [x for x in os.listdir('.') if os.path.isdir(x)]
dirs.sort()
odir = os.getcwd()
for d in dirs:
os.chdir(odir)
os.chdir(d)
pdbs = glob.glob('*.pdb')
cmds = []
resn = int(d.split('_')[1][2:])
for pdb in pdbs:
matchf = args.head + '_' + General.changeExt(pdb, 'match')
if not os.path.isfile(matchf):
continue
pos = PDB.findPositionInPDB(pdb, resn)
# if output file is already there, skip the job
if os.path.isfile('nr'+args.id +'_'+matchf):
continue
cmd = ['python', selfbin + '/removeLocalRedundancy.py', '--m', matchf, '--cres', str(pos), '--id', args.id, '--outh', 'nr'+args.id]
if not args.db == None:
cmd.extend(['--db', args.db])
if args.conR:
conresn = General.getBase(pdb).split('_')[2][1:]
conpos = PDB.findPositionInPDB(pdb, conresn)
cmd.extend(['--conres', str(conpos)])
# this script create .env file from .cmap files
import os, sys, csv, re
import General, Fragment, PDB
if len(sys.argv) -1 <1:
print '[Usage] <.cmap files as arguments>'
exit(0)
cmap_files = sys.argv[1:]
headers = ['residue', 'AA', 'total_contact_degree', 'crowdness', 'phipsi', 'permanent_contacts']
for cmap in cmap_files:
outf = General.changeExt(cmap, 'env_dun')
if os.path.isfile(outf):
continue
residues = {}
cmap_fh = open(cmap)
out_fh = open(outf, 'w')
f_csv = csv.DictWriter(out_fh, headers, delimiter = '\t')
f_csv.writeheader()
for line in cmap_fh:
if line.startswith('contact'):
res1, res2, degree, resname1, resname2 = line.rstrip('\n').split()[1:]
# resnum1, resnum2 = res1.split(',')[1], res2.split(',')[1]
# if not (resnum1.isdigit() and resnum2.isdigit()):
# continue
if not res1 in residues:
par.add_argument('--o', required = True, help = 'the output files')
args = par.parse_args()
odir = os.getcwd()
aas = [x for x in PDB.a2aaa]
pairtable = {}
for i in aas:
for j in aas:
pairtable[i+'|'+j] = 0
ctable = {x : 0 for x in aas}
for l in open(args.l):
info = l.strip().split('/')
subdir, name = info[-2], info[-1]
cfile = General.changeExt(subdir + '/' + name, args.ext)
if not os.path.isfile(cfile):
continue
for ll in open(cfile):
# if ll.find('contact') != 0:
# continue
info2 = ll.strip().split()
cond, aa1, aa2 = [info2[args.coln[i]] for i in range(3)]
aa1, aa2 = PDB.t2s(aa1), PDB.t2s(aa2)
# optional, for sc cond
sc_cond = float(info2[-1])
if sc_cond > 0.01:
continue
if (float(cond) >= args.range[0]) and (float(cond) <= args.range[1]):
import os, sys, argparse, csv
import General
par = argparse.ArgumentParser()
par.add_argument('--cmap', nargs = '+', help = '.cmap files, one or more')
args = par.parse_args()
for cmapf in args.cmap:
assert os.path.isfile(cmapf)
envf = General.changeExt(cmapf, 'env')
assert os.path.isfile(envf)
outf = General.changeExt(cmapf, 'cmapm')
total_cd = {}
with open(envf) as ef:
f_csv = csv.DictReader(ef, delimiter = '\t')
for row in f_csv:
resid, tcd = row['residue'], float(row['total_contact_degree'])
total_cd[resid] = tcd
out = open(outf, 'w')
for l in open(cmapf):
if not l.startswith('contact'):
continue
items = l.strip().split()
res1, res2, cd = items[1:4]
cd = round(float(cd), 3)
if (cd == 0) or (not res1 in total_cd) or (not res2 in total_cd):
continue
mutual_cd = cd * (0.5/total_cd[res1] + 0.5/total_cd[res2])
nitems = [x for x in items]
import General, PDB
par = argparse.ArgumentParser()
par.add_argument('--l', required = True, help = 'the used list file')
par.add_argument('--e', default = 'env', help = 'the extension of files with environment score')
par.add_argument('--c', default = 'cmap', help = 'the extension of files with contacts')
par.add_argument('--range', nargs = 2, default = [0.01, 1], help = 'the range of contact degree')
par.add_argument('--o', required = True, help = 'the output file')
args = par.parse_args()
out = open(args.o, 'w')
for l in open(args.l):
info = l.strip().split('/')
subdir, name = info[-2], info[-1]
# use the environment file for the whole protein but cmap file for single chain, don't know if this is good
envfile = General.changeExt(subdir + '/' + name.split('_')[0], args.e)
cmapfile = General.changeExt(subdir + '/' + name, args.c)
if not (os.path.isfile(envfile) and os.path.isfile(cmapfile)):
continue
env = {}
with open(envfile) as ef:
f_csv = csv.DictReader(ef, delimiter = '\t')
for row in f_csv:
env[row['residue']] = row['environment_score']
cf = open(cmapfile)
for cfl in cf:
if not cfl.startswith('contact'):
continue
par.add_argument('--sl', help = 'use a searchDB list file')
par.add_argument('--o', required = True, help = 'name of the output file')
args = par.parse_args()
out = open(args.o, 'w')
def outputSeq(seqs, name, out, chains = None):
if (chains != None) and (not isinstance(chains, list)):
chains = list(chains)
keys = seqs.keys()
keys.sort()
for k in keys:
if (chains != None) and (k not in chains):
continue
out.write('>' + name + '_' + k + '\n')
out.write(seqs[k]+'\n')
if args.sl == None:
for l in open(args.pl):
pid, cid = l.strip().split('_')
p = pid.lower() + '.clean.pdb'
seqs = PDB.pdb2seq(p)
outputSeq(seqs, pid.lower(), out, cid)
else:
for l in open(args.sl):
p = General.changeExt(l.rstrip('\n'), 'pdb')
seqs = PDB.pdb2seq(p)
name = General.removePath(p).split('.')[0]
outputSeq(seqs, name, out)
out.close()
assert args.ppfile != None
pp_file = args.ppfile
pp_lines = open(pp_file).readlines()
pp_bins = [float(x) for x in pp_lines[0].strip().split()]
pp_aatypes = pp_lines[2].strip().split()
pp_aaindex = {pp_aatypes[x] : x for x in range(20)}
pp_pots = np.zeros((36, 36, 20))
for ppl in pp_lines[4:]:
ppl_items = ppl.strip().split()
pp_pots[int(ppl_items[0])-1, int(ppl_items[1])-1] = map(float, ppl_items[2:])
seqfs = glob.glob(args.head + '*.seq')
cid, resnum = args.resid
for seqf in seqfs:
pdbf = General.changeExt( seqf.replace(args.head + '_', ''), 'pdb')
if not os.path.isfile(pdbf):
print(pdbf + ' doesn\'t exist!')
continue
outf = General.changeExt(pdbf, args.o)
if args.wgap != None: # specific to gap
assert args.conR == False, 'wgap and conR cannot be specified simultaneously'
dirname = General.getBase(pdbf)
pdbf = args.wgap + '/' + dirname + '/'+ pdbf
index = PDB.findPositionInPDB(pdbf, resnum, cid)
aacol = Analyze.readColumn(seqf, index, top = args.uplimit)
if args.conR: # should contacting residue be constrained?
key=lambda x: (
x.split()[0].split(",")[0],
int(re.search("\d+", x.split()[0]).group(0)),
x.split()[1].split(",")[0],
int(re.search("\d+", x.split()[1]).group(0)),
),
)
for outstring in sort_outstrs:
out_fh.write(outstring + "\n")
out_fh.close()
for pdb in args.p:
assert os.path.isfile(pdb)
# run confind in verbose mode
cmapv = General.changeExt(pdb, "cmapv")
if not os.path.isfile(cmapv):
cmap_fh = open(cmapv, "w")
cmd_confind = [CONFIND, "--p", pdb, "--rLib", rotLib, "--verb"]
sub.call(cmd_confind, stdout=cmap_fh)
cmap_fh.close()
# parse cmapv file for all pairs of contacts
G, V = VerboseToGraph(cmapv, pdb)
outf = General.changeExt(pdb, args.ext)
if not args.var:
calculator(G, outf)
else:
calculator(G, outf, V)
os.remove(cmapv)
uplimit = args.uplimit
nseq = 0
for match_line in open(args.m):
if (uplimit != None) and (nseq == uplimit):
break
match_line = match_line.strip()
indices = Analyze.index_from_match(match_line)
index1, index2 = indices[args.n[0]], indices[args.n[1]]
target_pds = match_line.split()[1]
targetid = General.getBase( General.removePath(match_line.split()[1]) )
env_dict = database_path + '/' + targetid[1:3] + '/' + targetid + '.freedom.db'
db = shelve.open(env_dict, 'r')
# extract post-processed pdb files from target_pds
resfile = database_path + '/' + targetid[1:3] + '/' + General.changeExt( General.removePath(target_pds), 'post.res')
allres = open(resfile).read().splitlines()
resid1, resid2 = allres[index1], allres[index2]
resid1, resid2 = resid1[0] + ',' + resid1[1:], resid2[0] + ',' + resid2[1:]
fields = ['sumcond', 'crwdnes', 'freedom', 'phi', 'psi', 'aa']
outfh.write(targetid + '\t')
if not resid1 in db:
outfh.write('\t'.join([resid1] + ['NA' for x in range(len(fields))]))
else:
res_info = db[resid1]
outfields = [str(res_info[x]) if x in res_info else 'NA' for x in fields]
outfh.write('\t'.join([resid1] + outfields))
outfh.write('\t')
if not resid2 in db:
outfh.write('\t'.join([resid2] + ['NA' for x in range(len(fields))]))
else:
args = parser.parse_args()
# parameters required for using smart rmsd cutoff from Craig
# rmsdmax, perLen = 1,1, 15
# dependencies between arguments
if (args.rmsd == None) and (args.bbrmsd == None) and (args.nohomo == None) and (not args.uniq) and (not args.smart):
raise General.myerror('I am doing nothing...')
if (args.uniq == False) and (args.nonat == True):
raise General.myerror('cannot specify nonat without uniq')
if args.ohead == args.head:
raise General.myerror('after process the head name is the same, not allowed...')
pid = args.pdb.split('_')[0]
matchf = args.head+'_'+General.changeExt(args.pdb, 'match')
seqf = General.changeExt(matchf, 'seq')
conres = PDB.ConRes(args.pdb)
# if using smart rmsd cutoff, need to create a list in which each element is the length of a segment
if args.smart:
resnums = [r.getResnum() for r in conres]
resnums.sort()
segments = [1]
for i in range(1, len(resnums)):
if resnums[i] - 1 == resnums[i-1]:
segments[-1] += 1
else:
segments.append(1)
par.add_argument('--db', default = '/home/anthill/fzheng/home/searchDB/statistics/bc-30-sc-20141022.peprm2.db', help = 'a shelve db object which contains the sequences of database targets')
par.add_argument('--cres', required = True, type = int, help = 'the index of the central position in the match, start from 1')
par.add_argument('--wd', default = 15, type = int, help = 'the size of comparing window on each side of the central position, so 7 means a 15 residue window')
par.add_argument('--id', required = True, help = 'identity cutoff for clustering')
par.add_argument('--outh', default = 'nr', help = 'a head to put before processed seq and match file')
par.add_argument('--conres', type = int, help = 'the index of the contacting position')
par.add_argument('--env', default = 'envpair', help = 'the extension of the environment file')
args = par.parse_args()
matches = open(args.m)
database = shelve.open(args.db)
# create a temporary file of all sequence context
odir = os.getcwd()
ldir = General.createLocalSpace()
tempfile = General.changeExt(args.m, 'seqcontext.fasta')
tempfh = open(ldir + '/' + tempfile, 'w')
matchind = 0
# if consider the contacting position
if args.conres != None:
tempfile2 = General.changeExt(args.m, 'seqcontext.fasta2')
tempfh2 = open(ldir + '/' + tempfile2, 'w')
# output file names
nr_matchf = args.outh + '_' + args.m
nr_seqf = General.changeExt(nr_matchf, 'seq')
nr_env = None
oenv = General.changeExt(args.m, args.env)
if os.path.isfile(oenv):
nr_env = General.changeExt(nr_matchf, args.env)
import itertools
sys.path.insert(0, '/home/grigoryanlab/home/fzheng/modules_py')
import General
if len(sys.argv) - 1 != 2:
print '<Usage> [a .cmap file] [a threshold]'
cmap, cut = sys.argv[1:]
condict = {}
residues = {}
for l in open(cmap):
info = l.strip().split()
if l.startswith('contact') and (float(info[3]) > float(cut)):
residues[info[1]] = info[4]
residues[info[2]] = info[5]
condict[info[1]+'-'+info[2]] = info[3]
resi = sorted(residues.keys())
combinations = itertools.combinations(resi, 3)
out = open(General.changeExt(cmap,'con3'), 'w')
for cb in combinations:
r1, r2, r3 = cb
if (r1+'-'+r2 in condict) and (r1+'-'+r3 in condict) and (r2+'-'+r3 in condict):
out.write('\t'.join([r1, r2, r3, residues[r1], residues[r2], residues[r3],
condict[r1+'-'+r2], condict[r1+'-'+r3], condict[r2+'-'+r3]]) + '\n')
out.close()
errorfree = True
smart = False
if args.rmsd == None:
smart = True
else:
rmsdeff = args.rmsd
for pdb in args.p:
# set rmsd cutoff using Craig's function
residues = PDB.ConRes(pdb)
segments = Fragment.getSegments(residues)
if smart:
rmsdeff = mustpress.rmsdEff(segments, args.params[0], args.params[1])
seqout = args.head + '_' + General.changeExt(pdb, 'seq')
matchout = General.changeExt(seqout, 'match')
if os.path.isfile(seqout):
continue
# search is additonal to a previous round
if args.more != None:
oseqf = args.more + '_' + General.changeExt(pdb, 'seq')
if not os.path.isfile(oseqf):
continue
olen = sum([1 for x in open(oseqf)])
if olen >= args.topn:
omatchf = General.changeExt(oseqf, 'match')
os.system('ln -s ' + oseqf + ' ' + seqout)
os.system('ln -s ' + omatchf + ' ' + matchout)
pdbs = glob.glob(FRAGMENTS_OUT + '*.pdb')
pdbs.sort()
ldir = General.createLocalSpace()
for p in pdbs:
base = General.removePath( General.getBase(p) )
if base.startswith('hit1'):
odir = ABUNDANCE_OUT
seqf = ABUNDANCE_OUT + args.h[1] + '_' + base + '.seq'
else:
odir = DESIGNSCORE_OUT
seqf = DESIGNSCORE_OUT + args.h[0] + '_' + base + '.seq'
seqout = args.nh + '_' + base + '.seq'
matchout = General.changeExt(seqout, 'match')
if os.path.isfile(odir + seqout):
continue
nlines = 0
if os.path.isfile(seqf):
for l in open(seqf):
nlines += 1
if nlines == args.cut:
break
if nlines >= args.cut:
continue
pds_file = changeExt(p, 'pds')
cmd_master = [Master + 'master', '--query', pds_file, '--targetList', list_tmp, '--rmsdCut', args.rmsd, '--topN', str(args.cut), '--bbRMSD', '--matchOut', ldir + '/' + matchout, '--seqOut', ldir + '/' +seqout]
sub.call(cmd_master)
exit(0)
lst, out = sys.argv[1:]
# aa is in a certain order
aatypes = ['G', 'A', 'V', 'L', 'I', 'M', 'F', 'W', 'P', 'S', 'T', 'C', 'Y', 'N', 'Q', 'D', 'E', 'K', 'R', 'H']
aaindex = {}
for i in range(20):
aaindex[aatypes[i]] = i
# make an 20^3 matrix
mat = np.zeros((20,20,20), dtype=int)
for l in open(lst):
info = l.strip().split('/')
name, pds = info[-2], info[-1]
con3f = name + '/' + General.changeExt(pds, 'con3')
if not os.path.isfile(con3f):
continue
fh = open(con3f)
for ll in fh:
ll = ll.strip()
aas = ll.split()[3:6]
aas = map(PDB.t2s, aas)
permut = itertools.permutations(aas, 3)
for k in permut:
mat[aaindex[k[0]], aaindex[k[1]], aaindex[k[2]]] += 1
np.save(out, mat)
par = argparse.ArgumentParser()
par.add_argument('--f', required = True, help = 'a .tab file')
par.add_argument('--verb', action = 'store_true', help = 'if output verbose')
args = par.parse_args()
CONFIND = '/home/grigoryanlab/home/gevorg/work/MSL/latest/trunk/bin/confind'
rotLib = '/home/anthill/fzheng/home/scripts/termanal/support.default/rotlib/RR2000.rotlib'
for l in open(args.f):
l = l.rstrip('\n')
pid, cid = l.split()[0:2]
cleanpdb = pid.lower() + '.clean.pdb'
if not os.path.isfile(cleanpdb):
continue
if args.verb:
cmap = General.changeExt(cleanpdb,'cmapv')
else:
cmap = General.changeExt(cleanpdb,'cmap')
if os.path.isfile(cmap):
continue
with warnings.catch_warnings():
try:
if args.verb:
out = open(cmap, 'w')
cmd_confind = [CONFIND, '--p', cleanpdb, '--rLib', rotLib, '--verb']
sub.call(cmd_confind, stdout = out)
else:
cmd_confind = [CONFIND, '--p', cleanpdb, '--o', cmap, '--rLib', rotLib]
sub.call(cmd_confind)
except Warning:
print 'warnings in', pid
env_data_path = '/home/anthill/fzheng/home/searchDB/support_bc-30-sc-correct-20141022/others'
for res in residues:
cid, resnum = res.getChid().strip(), res.getResnum()
if not args.post:
frag_pdb = FRAGMENTS_OUT + '/' + base + '_' + cid + str(resnum) + '.pdb'
pds_file = changeExt(frag_pdb, 'pds')
seqf1 = DESIGNSCORE_OUT + '/' + head[1] + '_' + base + '_' + cid + str(resnum) + '.seq'
else:
frag_pdb = outdir + '/' + base + '/fragments/' + base + '_' + cid + str(resnum) + '.pdb'
pds_file = changeExt(frag_pdb, 'pds')
seqf1 = outdir + '/' + base + '/designscore/' + head[1] + '_' + base + '_' + cid + str(resnum) + '.seq'
if not os.path.isfile(frag_pdb):
continue
matchf1 = General.changeExt(seqf1, 'match')
cmd_matchInFile = [MASTER + '/master', '--query', pds_file, '--matchIn', matchf1, '--structOut', General.changeExt(seqf1, 'sout'), '--outType', 'match', '--bbRMSD', '--topN', str(topN)]
cmd_matchInFile = sub.call(cmd_matchInFile)
match_lines = []
nmatch = 0
for match_l in open(matchf1):
nmatch += 1
match_lines.append(match_l)
if nmatch == topN:
break
# determine the index of original residue in its own TERM
rn = findPositionInPDB(frag_pdb, resnum, cid)
envf = open(General.changeExt(seqf1, 'envi'), 'w')
for i in range(1, len(match_lines)+1):
midx = str(i).rjust(len(str(len(match_lines))), '0')
sys.path.insert(0, "/home/grigoryanlab/home/fzheng/modules_py/")
import General
par = argparse.ArgumentParser()
par.add_argument("--fa", required=True, help="a fasta file")
par.add_argument("--o", default="peprm", help="extension of the output file")
par.add_argument(
"--cut", default=30, help="the cutoff, when the sequence length is lower than this, it is removed from the list"
)
args = par.parse_args()
dir = os.path.dirname(os.path.realpath(sys.argv[0])) + "/../"
lines = open(args.fa).readlines()
# will have two output file, new list and new fa
out_l = open("list_" + args.o, "w")
out_fa = open(General.changeExt(args.fa, args.o + ".fa"), "w")
for i in range(0, len(lines), 2):
length = len(lines[i + 1].strip())
if length < args.cut:
continue
name = General.changeExt(lines[i].strip()[1:], "pds")
path = os.path.realpath(dir) + "/" + name[1:3] + "/" + name
out_l.write(path + "\n")
out_fa.write(lines[i])
out_fa.write(lines[i + 1])
out_l.close()
out_fa.close()