def coord_to_pdb(coords, sequence, coord_id, clean=True):
assert len(sequence) == len(coords)
pdbca = "{0}.pdb".format(coord_id)
if not pdb_from_ca(coords, sequence, pdbca):
raise RuntimeError("Error processing coords {0}".format(coord_id))
#print("Wrote file {0}".format(pdbca))
pdb_side = ample_util.filename_append(pdbca, "pulchra")
if not add_sidechains_pulchra(pdbca, pdb_side):
print "Failed to add pulchra sidechains for coord {0}".format(coord_id)
pdb_side = ample_util.filename_append(pdbca, "maxsprout")
if not add_sidechains_maxsprout(pdbca, pdb_side):
print "*** Failed to add maxsprout sidechains for coord {0}".format(
coord_id)
return None
# Now add sidechains with SCWRL
pdbout = ample_util.filename_append(pdb_side, "scwrl")
SCWRL.add_sidechains(pdbin=pdb_side,
pdbout=pdbout,
hydrogens=False,
strip_oxt=False)
if clean: map(os.unlink, [pdbca, pdb_side])
return pdbout
def split_into_chains(pdbin, chain=None, directory=None):
"""Split a pdb file into its separate chains"""
if directory is None:
directory = os.path.dirname(pdbin)
# Largely stolen from pdb_split_models.py in phenix
#http://cci.lbl.gov/cctbx_sources/iotbx/command_line/pdb_split_models.py
pdbf = iotbx.file_reader.any_file(pdbin, force_type="pdb")
pdbf.check_file_type("pdb")
hierarchy = pdbf.file_object.construct_hierarchy()
# Nothing to do
n_models = hierarchy.models_size()
if n_models != 1:
raise RuntimeError("split_into_chains only works with single-mdoel pdbs!")
crystal_symmetry = pdbf.file_object.crystal_symmetry()
output_files = []
n_chains = len(hierarchy.models()[0].chains())
for i, hchain in enumerate(hierarchy.models()[0].chains()):
if not hchain.is_protein():
continue
if chain and not hchain.id == chain:
continue
new_hierarchy = iotbx.pdb.hierarchy.root()
new_model = iotbx.pdb.hierarchy.model()
new_hierarchy.append_model((new_model))
new_model.append_chain(hchain.detached_copy())
output_file = ample_util.filename_append(pdbin, hchain.id, directory)
with open(output_file, "w") as f:
if (crystal_symmetry is not None):
f.write(
iotbx.pdb.format_cryst1_and_scale_records(
crystal_symmetry=crystal_symmetry, write_scale_records=True
) + '\n'
)
f.write('REMARK Chain %d of %d\n' % (i, n_chains))
if (pdbin is not None):
f.write('REMARK Original file:\n')
f.write('REMARK %s\n' % pdbin)
f.write(new_hierarchy.as_pdb_string())
output_files.append(output_file)
if not len(output_files):
raise RuntimeError("split_into_chains could not find any chains to split")
return output_files
def del_column(file_name, column, overwrite=True):
"""Delete a column from an mtz file and return a path to the file"""
mtzDel = ample_util.filename_append(file_name, "d{0}".format(column))
cmd = ["mtzutils", "hklin1", file_name, "hklout", mtzDel]
stdin = "EXCLUDE 1 {0}".format(column)
logfile = os.path.join(os.getcwd(), "mtzutils.log")
retcode = ample_util.run_command(cmd, stdin=stdin, logfile=logfile)
if retcode != 0:
msg = "Error running mtzutils. Check the logfile: {0}".format(logfile)
logger.critical(msg)
raise RuntimeError(msg)
if overwrite:
shutil.move(mtzDel, file_name)
return file_name
else:
return mtzDel
def add_rfree(file_name,directory=None,overwrite=True):
"""Run uniqueify on mtz file to generate RFREE data column"""
mtzUnique = ample_util.filename_append(file_name, "uniqueify", directory=directory)
cmd = ['uniqueify', file_name, mtzUnique]
logfile = os.path.join(os.getcwd(), "uniqueify_{}.log".format(str(uuid.uuid1())))
retcode = ample_util.run_command(cmd, logfile=logfile)
if retcode != 0:
raise RuntimeError("Error running command: {0}. Check the logfile: {1}".format(" ".join(cmd), logfile))
else:
os.unlink(logfile)
if overwrite:
shutil.move(mtzUnique,file_name)
return file_name
else:
return mtzUnique
def del_column(file_name, column, overwrite=True):
"""Delete a column from an mtz file and return a path to the file"""
mtzDel = ample_util.filename_append(file_name, "d{0}".format(column) )
cmd = [ "mtzutils", "hklin1", file_name, "hklout", mtzDel ]
stdin = "EXCLUDE 1 {0}".format( column )
logfile = os.path.join(os.getcwd(), "mtzutils_{}.log".format(str(uuid.uuid1())))
retcode = ample_util.run_command(cmd, stdin=stdin, logfile=logfile)
if retcode != 0:
raise RuntimeError("Error running mtzutils. Check the logfile: {0}".format(logfile))
else:
os.unlink(logfile)
if overwrite:
shutil.move(mtzDel,file_name)
return file_name
else:
return mtzDel
def add_rfree(file_name, directory=None, overwrite=True):
"""Run uniqueify on mtz file to generate RFREE data column"""
mtzUnique = ample_util.filename_append(file_name,
"uniqueify",
directory=directory)
cmd = ['uniqueify', file_name, mtzUnique]
logfile = os.path.join(os.getcwd(), "uniqueify.log")
retcode = ample_util.run_command(cmd, logfile=logfile)
if retcode != 0:
msg = "Error running command: {0}. Check the logfile: {1}".format(
" ".join(cmd), logfile)
logger.critical(msg)
raise RuntimeError(msg)
if overwrite:
shutil.move(mtzUnique, file_name)
return file_name
else:
return mtzUnique
def split_pdb(pdbin, directory=None, strip_hetatm=False, same_size=False):
"""Split a pdb file into its separate models
Parameters
----------
pdbin : str
path to input pdbf file
directory : str
path to directory where pdb files will be created
strip_hetatm : bool
remove HETATMS if true
same_size : bool
Only output models of equal length (the most numerous length is selected)
"""
if directory is None:
directory = os.path.dirname(pdbin)
if not os.path.isdir(directory):
os.mkdir(directory)
# Largely stolen from pdb_split_models.py in phenix
#http://cci.lbl.gov/cctbx_sources/iotbx/command_line/pdb_split_models.py
pdbf = iotbx.file_reader.any_file(pdbin, force_type="pdb")
pdbf.check_file_type("pdb")
hierarchy = pdbf.file_object.construct_hierarchy()
# Nothing to do
n_models = hierarchy.models_size()
if same_size:
_only_equal_sizes(hierarchy)
crystal_symmetry = pdbf.file_object.crystal_symmetry()
output_files = []
for k, model in enumerate(hierarchy.models()):
k += 1
new_hierarchy = iotbx.pdb.hierarchy.root()
new_hierarchy.append_model(model.detached_copy())
if strip_hetatm: _strip(
new_hierarchy,
hetatm=True,
)
if (model.id == ""):
model_id = str(k)
else:
model_id = model.id.strip()
output_file = ample_util.filename_append(pdbin, model_id, directory)
with open(output_file, "w") as f:
if (crystal_symmetry is not None):
f.write(
iotbx.pdb.format_cryst1_and_scale_records(
crystal_symmetry=crystal_symmetry, write_scale_records=True
) + '\n'
)
f.write("REMARK Model %d of %d\n" % (k, n_models))
if (pdbin is not None):
f.write('REMARK Original file:\n')
f.write('REMARK %s\n' % pdbin)
f.write(new_hierarchy.as_pdb_string())
output_files.append(output_file)
return output_files
import os
import sys
sys.path.append("/Users/jmht/Documents/AMPLE/ample-dev1/scripts")
sys.path.append("/Users/jmht/Documents/AMPLE/ample-dev1/python")
import ample_util
from analyse_run import AmpleResult
ocsv = sys.argv[1]
ncsv = ample_util.filename_append( filename=ocsv, astr="obj" )
a = AmpleResult()
with open( ocsv ) as o, open( ncsv, 'w' ) as n:
for i, line in enumerate( o ):
if i == 0:
fields = line.strip().split(",")
nf = []
for f in fields:
i = a.orderedTitles.index( f )
nf.append( a.orderedAttrs[ i ] )
line = ",".join( nf ) + "\n"
n.write( line )
print "processing ", r.pdbCode, r.ensembleName
#r.orderedAttrs = a.orderedAttrs
#r.orderedTitles = a.orderedTitles
buccaneerLog = os.path.join(runDir, r.pdbCode, "ROSETTA_MR_0", "MRBUMP",
"cluster_1", "search_{0}_mrbump".format(
r.ensembleName), "data",
"loc0_ALL_{0}".format(r.ensembleName), "unmod",
"mr", "{0}".format(r.mrProgram), "build",
"shelxe", "rebuild", "build", "buccaneer.log")
if os.path.isfile(buccaneerLog):
p.parse(buccaneerLog)
r.buccFinalRfact = p.finalRfact
r.buccFinalRfree = p.finalRfree
pfile = ample_util.filename_append(pfile, astr="bucc")
f = open(pfile, 'w')
ampleDict = cPickle.dump(allResults, f)
cpath = os.path.join(workDir, 'results_bucc.csv')
csvfile = open(cpath, 'wb')
csvwriter = csv.writer(csvfile,
delimiter=',',
quotechar='"',
quoting=csv.QUOTE_MINIMAL)
header = False
for r in allResults:
if not header:
#csvwriter.writerow( r.titlesAsList() )
csvwriter.writerow(r.valueAttrAsList())
print "\nResults for ", pdbCode
# Directory where all the data for this run live
dataDir = os.path.join(dataRoot, pdbCode)
# Get path to native Extract all the nativePdbInfo from it
nativePdb = os.path.join(dataDir, "{0}.pdb".format(pdbCode))
pdbedit = pdb_edit.PDBEdit()
nativePdbInfo = pdbedit.get_info(nativePdb)
# First check if the native has > 1 model and extract the first if so
if len(nativePdbInfo.models) > 1:
print "nativePdb has > 1 model - using first"
nativePdb1 = ample_util.filename_append(filename=nativePdb,
astr="model1",
directory=workdir)
pdbedit.extract_model(nativePdb,
nativePdb1,
modelID=nativePdbInfo.models[0].serial)
nativePdb = nativePdb1
# Standardise the PDB to rename any non-standard AA, remove solvent etc
nativePdbStd = ample_util.filename_append(filename=nativePdb,
astr="std",
directory=workdir)
pdbedit.standardise(nativePdb, nativePdbStd)
nativePdb = nativePdbStd
# Get the new Info about the native
nativePdbInfo = pdbedit.get_info(nativePdb)
# if os.path.isfile( phaserLog ):
# phaserP = phaser_parser.PhaserLogParser( phaserLog )
# ptime = phaserP.time
shelxeLog = os.path.join(mrDir, "build/shelxe/shelxe_run.log")
stime = 0.0
if os.path.isfile(shelxeLog):
shelxeP = parse_shelxe.ShelxeLogParser(shelxeLog)
stime = shelxeP.cputime
r.shelxeTime = stime
#print "PTIME ",r.phaserTime
#print "STIME ",stime
pfile = ample_util.filename_append(pfile, astr="timings")
f = open(pfile, 'w')
ampleDict = cPickle.dump(ensembleResults, f)
cpath = os.path.join(runDir, 'final_results_timings.csv')
csvfile = open(cpath, 'wb')
csvwriter = csv.writer(csvfile,
delimiter=',',
quotechar='"',
quoting=csv.QUOTE_MINIMAL)
header = False
for r in ensembleResults:
if not header:
#csvwriter.writerow( r.titlesAsList() )
csvwriter.writerow(r.valueAttrAsList())