def helixFromPdbs(self,
origin,
mrPdb,
nativePdb,
nativeChain,
dsspLog,
workdir=os.getcwd()):
"""This is a wrapper to generate the info and resSeqMap objects needed by score Origin"""
mrPdbInfo = pdb_edit.get_info(mrPdb)
nativePdbInfo = pdb_edit.get_info(nativePdb)
assert nativeChain in nativePdbInfo.models[0].chains
import residue_map
resSeqMap = residue_map.residueSequenceMap()
resSeqMap.fromInfo(
refInfo=nativePdbInfo,
refChainID=nativeChain,
targetInfo=mrPdbInfo,
targetChainID=mrPdbInfo.models[0].chains[
0] # Only 1 chain in model
)
data = self.scoreOrigin(origin, mrPdbInfo, nativePdbInfo, resSeqMap,
workdir)
contacts = data.contacts
return self.helixFromContacts(contacts, dsspLog)
def analyseModels(amoptd):
# Get hold of a full model so we can do the mapping of residues
refModelPdb = glob.glob(os.path.join(amoptd['models_dir'], "*.pdb"))[0]
nativePdbInfo=amoptd['native_pdb_info']
refModelPdbInfo = pdb_edit.get_info(refModelPdb)
amoptd['ref_model_pdb_info']=refModelPdbInfo
try:
resSeqMap = residue_map.residueSequenceMap()
resSeqMap.fromInfo( refInfo=refModelPdbInfo,
refChainID=refModelPdbInfo.models[0].chains[0], # Only 1 chain in model
targetInfo=nativePdbInfo,
targetChainID=nativePdbInfo.models[0].chains[0]
)
amoptd['res_seq_map'] = resSeqMap
except Exception as e:
logger.exception("Error calculating resSeqMap: %s" % e)
amoptd['res_seq_map'] = None # Won't be able to calculate RIO scores
if amoptd['have_tmscore']:
try:
tm = tm_util.TMscore(amoptd['tmscore_exe'], wdir=fixpath(amoptd['benchmark_dir']))
# Calculation of TMscores for all models
logger.info("Analysing Rosetta models with TMscore")
model_list = sorted(glob.glob(os.path.join(amoptd['models_dir'], "*pdb")))
structure_list = [amoptd['native_pdb_std']]
amoptd['tmComp'] = tm.compare_structures(model_list, structure_list, fastas=[amoptd['fasta']])
except Exception as e:
logger.exception("Unable to run TMscores: %s", e)
else:
raise RuntimeError("No program to calculate TMSCORES")
def test_resSeqMap4(self):
# See if we can sort out the indexing between the native and model
nativePdb = os.path.join(self.testfiles_dir, "1K33.pdb")
modelPdb = os.path.join(self.testfiles_dir, "1K33_S_00000001.pdb")
nativePdbStd = "1K33_std.pdb"
pdb_edit.standardise(nativePdb, nativePdbStd)
nativeInfo = pdb_edit.get_info(nativePdbStd)
modelInfo = pdb_edit.get_info(modelPdb)
resSeqMap = residue_map.residueSequenceMap()
resSeqMap.fromInfo(nativeInfo, 'A', modelInfo, 'A')
os.unlink(nativePdbStd)
def test_resSeqMap4(self):
# See if we can sort out the indexing between the native and model
nativePdb = os.path.join(self.testfiles_dir,"1K33.pdb")
modelPdb = os.path.join(self.testfiles_dir,"1K33_S_00000001.pdb")
nativePdbStd = "1K33_std.pdb"
pdb_edit.standardise( nativePdb, nativePdbStd )
nativeInfo = pdb_edit.get_info( nativePdbStd )
modelInfo = pdb_edit.get_info( modelPdb )
resSeqMap = residue_map.residueSequenceMap( )
resSeqMap.fromInfo( nativeInfo, 'A', modelInfo, 'A' )
os.unlink( nativePdbStd )
def helixFromPdbs(self, origin, mrPdb, nativePdb, nativeChain, dsspLog, workdir=os.getcwd() ):
"""This is a wrapper to generate the info and resSeqMap objects needed by score Origin"""
mrPdbInfo = pdb_edit.get_info(mrPdb)
nativePdbInfo = pdb_edit.get_info(nativePdb)
assert nativeChain in nativePdbInfo.models[0].chains
import residue_map
resSeqMap = residue_map.residueSequenceMap()
resSeqMap.fromInfo( refInfo=nativePdbInfo,
refChainID=nativeChain,
targetInfo=mrPdbInfo,
targetChainID=mrPdbInfo.models[0].chains[0]# Only 1 chain in model
)
data = self.scoreOrigin(origin, mrPdbInfo, nativePdbInfo, resSeqMap, workdir)
contacts = data.contacts
return self.helixFromContacts(contacts, dsspLog )
def analyseModels(amoptd):
# Get hold of a full model so we can do the mapping of residues
refModelPdb = glob.glob(os.path.join(amoptd['models_dir'], "*.pdb"))[0]
nativePdbInfo = amoptd['native_pdb_info']
resSeqMap = residue_map.residueSequenceMap()
refModelPdbInfo = pdb_edit.get_info(refModelPdb)
resSeqMap.fromInfo(
refInfo=refModelPdbInfo,
refChainID=refModelPdbInfo.models[0].
chains[0], # Only 1 chain in model
targetInfo=nativePdbInfo,
targetChainID=nativePdbInfo.models[0].chains[0])
amoptd['res_seq_map'] = resSeqMap
amoptd['ref_model_pdb_info'] = refModelPdbInfo
if amoptd['have_tmscore']:
try:
tm = tm_util.TMscore(amoptd['tmscore_exe'],
wdir=fixpath(amoptd['benchmark_dir']))
# Calculation of TMscores for all models
logger.info("Analysing Rosetta models with TMscore")
model_list = sorted(
glob.glob(os.path.join(amoptd['models_dir'], "*pdb")))
structure_list = [amoptd['native_pdb_std']]
amoptd['tmComp'] = tm.compare_structures(model_list,
structure_list,
fastas=[amoptd['fasta']])
except Exception as e:
msg = "Unable to run TMscores: {0}".format(e)
logger.critical(msg)
else:
global _MAXCLUSTERER # setting a module-level variable so need to use global keyword to it doesn't become a local variable
_MAXCLUSTERER = maxcluster.Maxcluster(amoptd['maxcluster_exe'])
logger.info("Analysing Rosetta models with Maxcluster")
_MAXCLUSTERER.compareDirectory(nativePdbInfo=nativePdbInfo,
resSeqMap=resSeqMap,
modelsDirectory=amoptd['models_dir'],
workdir=fixpath(
amoptd['benchmark_dir']))
return
def analyseModels(amoptd):
# Get hold of a full model so we can do the mapping of residues
refModelPdb = glob.glob(os.path.join(amoptd['models_dir'], "*.pdb"))[0]
nativePdbInfo = amoptd['native_pdb_info']
refModelPdbInfo = pdb_edit.get_info(refModelPdb)
amoptd['ref_model_pdb_info'] = refModelPdbInfo
try:
resSeqMap = residue_map.residueSequenceMap()
resSeqMap.fromInfo(
refInfo=refModelPdbInfo,
refChainID=refModelPdbInfo.models[0].
chains[0], # Only 1 chain in model
targetInfo=nativePdbInfo,
targetChainID=nativePdbInfo.models[0].chains[0],
)
amoptd['res_seq_map'] = resSeqMap
except Exception as e:
logger.exception("Error calculating resSeqMap: %s" % e)
amoptd['res_seq_map'] = None # Won't be able to calculate RIO scores
if amoptd['have_tmscore']:
try:
tm = tm_util.TMscore(amoptd['tmscore_exe'],
wdir=fixpath(amoptd['benchmark_dir']))
# Calculation of TMscores for all models
logger.info("Analysing Rosetta models with TMscore")
model_list = sorted(
glob.glob(os.path.join(amoptd['models_dir'], "*pdb")))
structure_list = [amoptd['native_pdb_std']]
amoptd['tmComp'] = tm.compare_structures(model_list,
structure_list,
fastas=[amoptd['fasta']])
except Exception as e:
logger.exception("Unable to run TMscores: %s", e)
else:
raise RuntimeError("No program to calculate TMSCORES")
def analyseSolution(amoptd, d, mrinfo):
logger.info("Benchmark: analysing result: {0}".format(d['ensemble_name']))
mrPdb = None
if d['MR_program'] == "PHASER":
mrPdb = d['PHASER_pdbout']
mrMTZ = d['PHASER_mtzout']
elif d['MR_program'] == "MOLREP":
mrPdb = d['MOLREP_pdbout']
elif d['MR_program'] == "unknown":
return
if mrPdb is None or not os.path.isfile(mrPdb):
#logger.critical("Cannot find mrPdb {0} for solution {1}".format(mrPdb,d))
return
# debug - copy into work directory as reforigin struggles with long pathnames
shutil.copy(
mrPdb,
os.path.join(fixpath(amoptd['benchmark_dir']),
os.path.basename(mrPdb)))
mrPdbInfo = pdb_edit.get_info(mrPdb)
d['num_placed_chains'] = mrPdbInfo.numChains()
d['num_placed_atoms'] = mrPdbInfo.numAtoms()
d['num_placed_CA'] = mrPdbInfo.numCalpha()
if amoptd['native_pdb']:
if not d['SHELXE_os']:
logger.critical(
"mrPdb {0} has no SHELXE_os origin shift. Calculating...".
format(mrPdb))
mrinfo.analyse(mrPdb)
mrOrigin = mrinfo.originShift
d['SHELXE_MPE'] = mrinfo.MPE
d['SHELXE_wMPE'] = mrinfo.wMPE
else:
mrOrigin = [c * -1 for c in d['SHELXE_os']]
# Move pdb onto new origin
originPdb = ample_util.filename_append(mrPdb,
astr='offset',
directory=fixpath(
amoptd['benchmark_dir']))
#print(mrPdb, originPdb, mrOrigin)
pdb_edit.translate(mrPdb, originPdb, mrOrigin)
# offset.pdb is the mrModel shifted onto the new origin use csymmatch to wrap onto native
csymmatch.Csymmatch().wrapModelToNative(
originPdb,
amoptd['native_pdb'],
csymmatchPdb=os.path.join(
fixpath(amoptd['benchmark_dir']),
"phaser_{0}_csymmatch.pdb".format(d['ensemble_name'])))
# can now delete origin pdb
os.unlink(originPdb)
# Calculate phase error for the MR PDB
try:
mrinfo.analyse(mrPdb)
d['MR_MPE'] = mrinfo.MPE
d['MR_wMPE'] = mrinfo.wMPE
except Exception as e:
logger.critical("Error analysing mrPdb: {0}\n{1}".format(mrPdb, e))
# We cannot calculate the Reforigin RMSDs or RIO scores for runs where we don't have a full initial model
# to compare to the native to allow us to determine which parts of the ensemble correspond to which parts of
# the native structure.
if not (amoptd['homologs'] or \
amoptd['ideal_helices'] or \
amoptd['import_ensembles'] or \
amoptd['single_model_mode']):
# Get reforigin info
rmsder = reforigin.ReforiginRmsd()
try:
rmsder.getRmsd(nativePdbInfo=amoptd['native_pdb_info'],
placedPdbInfo=mrPdbInfo,
refModelPdbInfo=amoptd['ref_model_pdb_info'],
cAlphaOnly=True,
workdir=fixpath(amoptd['benchmark_dir']))
d['reforigin_RMSD'] = rmsder.rmsd
except Exception as e:
logger.critical("Error calculating RMSD: {0}".format(e))
d['reforigin_RMSD'] = 999
# Score the origin with all-atom and rio
rioData = rio.Rio().scoreOrigin(
mrOrigin,
mrPdbInfo=mrPdbInfo,
nativePdbInfo=amoptd['native_pdb_info'],
resSeqMap=amoptd['res_seq_map'],
workdir=fixpath(amoptd['benchmark_dir']))
# Set attributes
d['AA_num_contacts'] = rioData.aaNumContacts
d['RIO_num_contacts'] = rioData.rioNumContacts
d['RIO_in_register'] = rioData.rioInRegister
d['RIO_oo_register'] = rioData.rioOoRegister
d['RIO_backwards'] = rioData.rioBackwards
d['RIO'] = rioData.rioInRegister + rioData.rioOoRegister
d['RIO_no_cat'] = rioData.rioNumContacts - (rioData.rioInRegister +
rioData.rioOoRegister)
d['RIO_norm'] = float(d['RIO']) / float(
d['native_pdb_num_residues'])
else:
d['AA_num_contacts'] = None
d['RIO_num_contacts'] = None
d['RIO_in_register'] = None
d['RIO_oo_register'] = None
d['RIO_backwards'] = None
d['RIO'] = None
d['RIO_no_cat'] = None
d['RIO_norm'] = None
# # Now get the helix
# helixSequence = contacts.Rio().helixFromContacts( contacts=rioData.contacts,
# dsspLog=dsspLog )
# if helixSequence is not None:
# ampleResult.rioHelixSequence = helixSequence
# ampleResult.rioLenHelix = len( helixSequence )
# hfile = os.path.join( workdir, "{0}.helix".format( ampleResult.ensembleName ) )
# with open( hfile, 'w' ) as f:
# f.write( helixSequence+"\n" )
#
# This purely for checking and so we have pdbs to view
#
# Wrap shelxe trace onto native using Csymmatch
if not d['SHELXE_pdbout'] is None and os.path.isfile(
fixpath(d['SHELXE_pdbout'])):
csymmatch.Csymmatch().wrapModelToNative(
fixpath(d['SHELXE_pdbout']),
amoptd['native_pdb'],
origin=mrOrigin,
workdir=fixpath(amoptd['benchmark_dir']))
if not ('SHELXE_wMPE' in d and d['SHELXE_wMPE']):
try:
mrinfo.analyse(d['SHELXE_pdbout'])
d['SHELXE_MPE'] = mrinfo.MPE
d['SHELXE_wMPE'] = mrinfo.wMPE
except Exception as e:
logger.critical(
"Error analysing SHELXE_pdbout: {0}\n{1}".format(
d['SHELXE_pdbout'], e))
# Wrap parse_buccaneer model onto native
if d['SXRBUCC_pdbout'] and os.path.isfile(fixpath(
d['SXRBUCC_pdbout'])):
# Need to rename Pdb as is just called buccSX_output.pdb
csymmatchPdb = os.path.join(
fixpath(amoptd['benchmark_dir']),
"buccaneer_{0}_csymmatch.pdb".format(d['ensemble_name']))
csymmatch.Csymmatch().wrapModelToNative(
fixpath(d['SXRBUCC_pdbout']),
amoptd['native_pdb'],
origin=mrOrigin,
csymmatchPdb=csymmatchPdb,
workdir=fixpath(amoptd['benchmark_dir']))
# Calculate phase error
try:
mrinfo.analyse(d['SXRBUCC_pdbout'])
d['SXRBUCC_MPE'] = mrinfo.MPE
d['SXRBUCC_wMPE'] = mrinfo.wMPE
except Exception as e:
logger.critical(
"Error analysing SXRBUCC_pdbout: {0}\n{1}".format(
d['SXRBUCC_pdbout'], e))
# Wrap parse_buccaneer model onto native
if d['SXRARP_pdbout'] and os.path.isfile(fixpath(d['SXRARP_pdbout'])):
# Need to rename Pdb as is just called buccSX_output.pdb
csymmatchPdb = os.path.join(
fixpath(amoptd['benchmark_dir']),
"arpwarp_{0}_csymmatch.pdb".format(d['ensemble_name']))
csymmatch.Csymmatch().wrapModelToNative(
fixpath(d['SXRARP_pdbout']),
amoptd['native_pdb'],
origin=mrOrigin,
csymmatchPdb=csymmatchPdb,
workdir=fixpath(amoptd['benchmark_dir']))
# Calculate phase error
try:
mrinfo.analyse(d['SXRARP_pdbout'])
d['SXRARP_MPE'] = mrinfo.MPE
d['SXRARP_wMPE'] = mrinfo.wMPE
except Exception as e:
logger.critical(
"Error analysing SXRARP_pdbout: {0}\n{1}".format(
d['SXRARP_pdbout'], e))
return
def analysePdb(amoptd):
"""Collect data on the native pdb structure"""
nativePdb = fixpath(amoptd['native_pdb'])
nativePdbInfo = pdb_edit.get_info(nativePdb)
# number atoms/residues
natoms, nresidues = pdb_edit.num_atoms_and_residues(nativePdb)
# Get information on the origins for this spaceGroup
try:
originInfo = pdb_model.OriginInfo(
spaceGroupLabel=nativePdbInfo.crystalInfo.spaceGroup)
except Exception:
originInfo = None
# Do this here as a bug in pdbcur can knacker the CRYST1 data
amoptd['native_pdb_code'] = nativePdbInfo.pdbCode
amoptd['native_pdb_title'] = nativePdbInfo.title
amoptd['native_pdb_resolution'] = nativePdbInfo.resolution
amoptd['native_pdb_solvent_content'] = nativePdbInfo.solventContent
amoptd[
'native_pdb_matthews_coefficient'] = nativePdbInfo.matthewsCoefficient
if not originInfo:
space_group = "P1"
else:
space_group = originInfo.spaceGroup()
amoptd['native_pdb_space_group'] = space_group
amoptd['native_pdb_num_atoms'] = natoms
amoptd['native_pdb_num_residues'] = nresidues
# First check if the native has > 1 model and extract the first if so
if len(nativePdbInfo.models) > 1:
logger.info("nativePdb has > 1 model - using first")
nativePdb1 = ample_util.filename_append(filename=nativePdb,
astr="model1",
directory=fixpath(
amoptd['work_dir']))
pdb_edit.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=fixpath(
amoptd['work_dir']))
pdb_edit.standardise(nativePdb, nativePdbStd, del_hetatm=True)
nativePdb = nativePdbStd
# Get the new Info about the native
nativePdbInfo = pdb_edit.get_info(nativePdb)
# For maxcluster comparsion of shelxe model we need a single chain from the native so we get this here
if len(nativePdbInfo.models[0].chains) > 1:
chainID = nativePdbInfo.models[0].chains[0]
nativeChain1 = ample_util.filename_append(filename=nativePdbInfo.pdb,
astr="chain1",
directory=fixpath(
amoptd['work_dir']))
pdb_edit.to_single_chain(nativePdbInfo.pdb, nativeChain1)
else:
nativeChain1 = nativePdbInfo.pdb
# Additional data
amoptd['native_pdb_num_chains'] = len(nativePdbInfo.models[0].chains)
amoptd['native_pdb_info'] = nativePdbInfo
amoptd['native_pdb_std'] = nativePdbStd
amoptd['native_pdb_1chain'] = nativeChain1
amoptd['native_pdb_origin_info'] = originInfo
return
def analyseSolution(amoptd, d, mrinfo):
logger.info("Benchmark: analysing result: {0}".format(d['ensemble_name']))
mrPdb=None
if d['MR_program']=="PHASER":
mrPdb = d['PHASER_pdbout']
mrMTZ = d['PHASER_mtzout']
elif d['MR_program']=="MOLREP":
mrPdb = d['MOLREP_pdbout']
elif d['MR_program']=="unknown":
return
if mrPdb is None or not os.path.isfile(mrPdb):
#logger.critical("Cannot find mrPdb {0} for solution {1}".format(mrPdb,d))
return
# debug - copy into work directory as reforigin struggles with long pathnames
shutil.copy(mrPdb, os.path.join(fixpath(amoptd['benchmark_dir']), os.path.basename(mrPdb)))
mrPdbInfo = pdb_edit.get_info( mrPdb )
d['num_placed_chains'] = mrPdbInfo.numChains()
d['num_placed_atoms'] = mrPdbInfo.numAtoms()
d['num_placed_CA'] = mrPdbInfo.numCalpha()
if amoptd['native_pdb']:
if not d['SHELXE_os']:
logger.critical("mrPdb {0} has no SHELXE_os origin shift. Calculating...".format(mrPdb))
mrinfo.analyse(mrPdb)
mrOrigin = mrinfo.originShift
d['SHELXE_MPE'] = mrinfo.MPE
d['SHELXE_wMPE'] = mrinfo.wMPE
else:
mrOrigin=[c*-1 for c in d['SHELXE_os']]
# Move pdb onto new origin
originPdb = ample_util.filename_append(mrPdb, astr='offset',directory=fixpath(amoptd['benchmark_dir']))
#print(mrPdb, originPdb, mrOrigin)
pdb_edit.translate(mrPdb, originPdb, mrOrigin)
# offset.pdb is the mrModel shifted onto the new origin use csymmatch to wrap onto native
csymmatch.Csymmatch().wrapModelToNative(originPdb,
amoptd['native_pdb'],
csymmatchPdb=os.path.join(fixpath(amoptd['benchmark_dir']),
"phaser_{0}_csymmatch.pdb".format(d['ensemble_name'])))
# can now delete origin pdb
os.unlink(originPdb)
# Calculate phase error for the MR PDB
try:
mrinfo.analyse(mrPdb)
d['MR_MPE'] = mrinfo.MPE
d['MR_wMPE'] = mrinfo.wMPE
except Exception as e:
logger.critical("Error analysing mrPdb: {0}\n{1}".format(mrPdb,e))
# We cannot calculate the Reforigin RMSDs or RIO scores for runs where we don't have a full initial model
# to compare to the native to allow us to determine which parts of the ensemble correspond to which parts of
# the native structure - or if we were unable to calculate a res_seq_map
if not (amoptd['homologs'] or \
amoptd['ideal_helices'] or \
amoptd['import_ensembles'] or \
amoptd['single_model_mode'] or \
amoptd['res_seq_map']):
# Get reforigin info
rmsder = reforigin.ReforiginRmsd()
try:
rmsder.getRmsd(nativePdbInfo=amoptd['native_pdb_info'],
placedPdbInfo=mrPdbInfo,
refModelPdbInfo=amoptd['ref_model_pdb_info'],
cAlphaOnly=True,
workdir=fixpath(amoptd['benchmark_dir']))
d['reforigin_RMSD'] = rmsder.rmsd
except Exception as e:
logger.critical("Error calculating RMSD: {0}".format(e))
d['reforigin_RMSD'] = 999
# Score the origin with all-atom and rio
rioData = rio.Rio().scoreOrigin(mrOrigin,
mrPdbInfo=mrPdbInfo,
nativePdbInfo=amoptd['native_pdb_info'],
resSeqMap=amoptd['res_seq_map'],
workdir=fixpath(amoptd['benchmark_dir'])
)
# Set attributes
d['AA_num_contacts'] = rioData.aaNumContacts
d['RIO_num_contacts'] = rioData.rioNumContacts
d['RIO_in_register'] = rioData.rioInRegister
d['RIO_oo_register'] = rioData.rioOoRegister
d['RIO_backwards'] = rioData.rioBackwards
d['RIO'] = rioData.rioInRegister + rioData.rioOoRegister
d['RIO_no_cat'] = rioData.rioNumContacts - ( rioData.rioInRegister + rioData.rioOoRegister )
d['RIO_norm'] = float(d['RIO']) / float(d['native_pdb_num_residues'])
else:
d['AA_num_contacts'] = None
d['RIO_num_contacts'] = None
d['RIO_in_register'] = None
d['RIO_oo_register'] = None
d['RIO_backwards'] = None
d['RIO'] = None
d['RIO_no_cat'] = None
d['RIO_norm'] = None
# # Now get the helix
# helixSequence = contacts.Rio().helixFromContacts( contacts=rioData.contacts,
# dsspLog=dsspLog )
# if helixSequence is not None:
# ampleResult.rioHelixSequence = helixSequence
# ampleResult.rioLenHelix = len( helixSequence )
# hfile = os.path.join( workdir, "{0}.helix".format( ampleResult.ensembleName ) )
# with open( hfile, 'w' ) as f:
# f.write( helixSequence+"\n" )
#
# This purely for checking and so we have pdbs to view
#
# Wrap shelxe trace onto native using Csymmatch
if not d['SHELXE_pdbout'] is None and os.path.isfile(fixpath(d['SHELXE_pdbout'])):
csymmatch.Csymmatch().wrapModelToNative( fixpath(d['SHELXE_pdbout']),
amoptd['native_pdb'],
origin=mrOrigin,
workdir=fixpath(amoptd['benchmark_dir']))
if not('SHELXE_wMPE' in d and d['SHELXE_wMPE']):
try:
mrinfo.analyse(d['SHELXE_pdbout'])
d['SHELXE_MPE'] = mrinfo.MPE
d['SHELXE_wMPE'] = mrinfo.wMPE
except Exception as e:
logger.critical("Error analysing SHELXE_pdbout: {0}\n{1}".format(d['SHELXE_pdbout'],e))
# Wrap parse_buccaneer model onto native
if d['SXRBUCC_pdbout'] and os.path.isfile(fixpath(d['SXRBUCC_pdbout'])):
# Need to rename Pdb as is just called buccSX_output.pdb
csymmatchPdb = os.path.join(fixpath(amoptd['benchmark_dir']), "buccaneer_{0}_csymmatch.pdb".format(d['ensemble_name']))
csymmatch.Csymmatch().wrapModelToNative( fixpath(d['SXRBUCC_pdbout']),
amoptd['native_pdb'],
origin=mrOrigin,
csymmatchPdb=csymmatchPdb,
workdir=fixpath(amoptd['benchmark_dir']))
# Calculate phase error
try:
mrinfo.analyse(d['SXRBUCC_pdbout'])
d['SXRBUCC_MPE'] = mrinfo.MPE
d['SXRBUCC_wMPE'] = mrinfo.wMPE
except Exception as e:
logger.critical("Error analysing SXRBUCC_pdbout: {0}\n{1}".format(d['SXRBUCC_pdbout'],e))
# Wrap parse_buccaneer model onto native
if d['SXRARP_pdbout'] and os.path.isfile(fixpath(d['SXRARP_pdbout'])):
# Need to rename Pdb as is just called buccSX_output.pdb
csymmatchPdb = os.path.join(fixpath(amoptd['benchmark_dir']), "arpwarp_{0}_csymmatch.pdb".format(d['ensemble_name']))
csymmatch.Csymmatch().wrapModelToNative( fixpath(d['SXRARP_pdbout']),
amoptd['native_pdb'],
origin=mrOrigin,
csymmatchPdb=csymmatchPdb,
workdir=fixpath(amoptd['benchmark_dir']))
# Calculate phase error
try:
mrinfo.analyse(d['SXRARP_pdbout'])
d['SXRARP_MPE'] = mrinfo.MPE
d['SXRARP_wMPE'] = mrinfo.wMPE
except Exception as e:
logger.critical("Error analysing SXRARP_pdbout: {0}\n{1}".format(d['SXRARP_pdbout'],e))
return
def analysePdb(amoptd):
"""Collect data on the native pdb structure"""
nativePdb = fixpath(amoptd['native_pdb'])
nativePdbInfo = pdb_edit.get_info(nativePdb)
# number atoms/residues
natoms, nresidues = pdb_edit.num_atoms_and_residues(nativePdb)
# Get information on the origins for this spaceGroup
try:
originInfo = pdb_model.OriginInfo(spaceGroupLabel=nativePdbInfo.crystalInfo.spaceGroup)
except Exception:
originInfo = None
# Do this here as a bug in pdbcur can knacker the CRYST1 data
amoptd['native_pdb_code'] = nativePdbInfo.pdbCode
amoptd['native_pdb_title'] = nativePdbInfo.title
amoptd['native_pdb_resolution'] = nativePdbInfo.resolution
amoptd['native_pdb_solvent_content'] = nativePdbInfo.solventContent
amoptd['native_pdb_matthews_coefficient'] = nativePdbInfo.matthewsCoefficient
if not originInfo:
space_group = "P1"
else:
space_group = originInfo.spaceGroup()
amoptd['native_pdb_space_group'] = space_group
amoptd['native_pdb_num_atoms'] = natoms
amoptd['native_pdb_num_residues'] = nresidues
# First check if the native has > 1 model and extract the first if so
if len( nativePdbInfo.models ) > 1:
logger.info("nativePdb has > 1 model - using first")
nativePdb1 = ample_util.filename_append( filename=nativePdb, astr="model1", directory=fixpath(amoptd['work_dir']))
pdb_edit.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=fixpath(amoptd['work_dir']))
pdb_edit.standardise(nativePdb, nativePdbStd, del_hetatm=True)
nativePdb = nativePdbStd
# Get the new Info about the native
nativePdbInfo = pdb_edit.get_info( nativePdb )
# For comparsion of shelxe model we need a single chain from the native so we get this here
if len( nativePdbInfo.models[0].chains ) > 1:
nativeChain1 = ample_util.filename_append( filename=nativePdbInfo.pdb,
astr="chain1",
directory=fixpath(amoptd['work_dir']))
pdb_edit.merge_chains(nativePdbInfo.pdb, nativeChain1)
else:
nativeChain1 = nativePdbInfo.pdb
# Additional data
amoptd['native_pdb_num_chains'] = len( nativePdbInfo.models[0].chains )
amoptd['native_pdb_info'] = nativePdbInfo
amoptd['native_pdb_std'] = nativePdbStd
amoptd['native_pdb_1chain'] = nativeChain1
amoptd['native_pdb_origin_info'] = originInfo
return
