def setChainID( m ):
"""
set chainID for Hex pdb files
"""
if options['id']:
id = T.toList( options['id'] )
cMap = m.chainMap()
for chain in range( m.lenChains() ):
idx = N0.nonzero( cMap == chain )
for i in idx:
m.atoms['chain_id'][i] = id[chain]
def setChainID(m):
"""
set chainID for Hex pdb files
"""
if options['id']:
id = T.toList(options['id'])
cMap = m.chainMap()
for chain in range(m.lenChains()):
idx = N0.nonzero(cMap == chain)
for i in idx:
m.atoms['chain_id'][i] = id[chain]
def __setChainID( self, m, ids ):
"""
set chaiID for Hex pdb files
@param m: model
@type m: PDBModel
@param ids: chain id, len(ids) == m.lenChains
@type ids: [str]
@return: m is changed directly
@rtype: PDBModel
"""
if ids:
ids = t.toList( ids )
cMap = m.chainMap()
for chain in range( m.lenChains() ):
idx = N.nonzero( cMap == chain )
for i in idx:
m.atoms['chain_id'][i] = ids[chain]
def rmsdLimitedClustering( tc, options, min_cluster=5, max_cluster=30,
rmsLimit=1.0 ):
"""
Cluster iteratively until the average of all clusters meet
the rmsd cutoff criteria.
"""
allowedAtoms = T.toList( options.get('a',[]) )
if allowedAtoms:
mask = tc.traj.ref.mask( lambda a: a['name'] in allowedAtoms )
else:
mask = selectedAtoms( tc.traj.ref )
n_cluster = tc.calcClusterNumber( min_clst=min_cluster,
max_clst=max_cluster,
rmsLimit=1.0 )
options['n'] = n_cluster
cluster( tc, options )
def rmsdLimitedClustering( tc, options, min_cluster=5, max_cluster=30,
rmsLimit=1.0 ):
"""
Cluster iteratively until the average of all clusters meet
the rmsd cutoff criteria.
"""
allowedAtoms = T.toList( options.get('a',[]) )
if allowedAtoms:
mask = tc.traj.ref.mask( lambda a: a['name'] in allowedAtoms )
else:
mask = selectedAtoms( tc.traj.ref )
n_cluster = tc.calcClusterNumber( min_clst=min_cluster,
max_clst=max_cluster,
rmsLimit=1.0 )
options['n'] = n_cluster
cluster( tc, options )
T.dump(d, T.absfile(f))
##########
## MAIN ##
default = {'o': '', 'p': 'surf dens'}
if len(sys.argv) < 2:
_use(default)
options = T.cmdDict(default)
if options.has_key('i'):
inLst = T.toList(options['i'])
inLst = [T.absfile(f) for f in inLst]
else:
inLst = []
sourceIn = T.absfile(options['s'])
sourceOut = None
if options.has_key('so'):
sourceOut = T.absfile(options['so'])
prefix = options['o']
print 'Preparing source ' + str(os.path.basename(sourceIn))\
+ ' -> ' + str(sourceOut)
if importas == module:
importas = ''
else:
importas = ' as ' + importas
return l.replace( 'from '+module+' import *', 'import '+module+importas)
options = t.cmdDict( {'m':'Numeric', 'as':'N', 'e':['Complex'] } )
module = options['m']
importas = options.get('as', module )
srcfiles = t.toList( options.get('i', None) )
exclude = t.toList( options['e'] )
try:
exec('import '+module)
for fname in srcfiles:
fname = t.absfile( fname )
shutil.copy( fname, fname + '_' )
methods = re_lst( module, exclude )
fold = open( fname + '_' )
if len (sys.argv) < 2:
_use( default )
options = T.cmdDict( default )
## where to run the calculation
base_folder = T.absfile( options['r'] )+'/'
## template gly-xxx-gly
template = os.path.abspath( options['i'] )
## label the result dictionaty files
label = '_' + options['l']
## mask to used to delete glycines
mask = [ int(i) for i in T.toList( options['mask'] )]
## create random prptides from template
#randomPeptides( template, base_folder )
## collect average surfaces
MS, AS, MS_sd, AS_sd = randomSurfaces( base_folder, label, mask )
## save dictionary with all 20 amino acids
T.dump( MS, base_folder + 'MS%s.dic'%label)
T.dump( AS, base_folder + 'AS%s.dic'%label )
T.dump( MS_sd, base_folder + 'MS_sd%s.dic'%label )
T.dump( AS_sd, base_folder + 'AS_sd%s.dic'%label )
##########
## MAIN ##
options = tools.cmdDict()
if not 'i' in options:
print """Re-create LocalPath in PDBModel or Complex from the original
file name (and the current environment variables).
relocalize.py -i |file1 file2 ..|
"""
sys.exit(0)
files = tools.toList( options['i'] )
repl = []
if 'r' in options:
relacements = tools.toList( options['r'] )
repl = [ r.split( ':' ) for r in relacements ]
for f in files:
try:
print "Re-localizing ", f
f = tools.absfile( f )
o = tools.load( f )
result = 0
if o.__class__ in [ PCRModel, PDBModel ]:
result = localizeModel( o, repl, f )
m.update( updateMissing=1 )
T.dump( d, T.absfile( f ) )
##########
## MAIN ##
default = {'o':'', 'p':'surf dens'}
if len (sys.argv) < 2:
_use( default )
options = T.cmdDict( default )
if options.has_key('i'):
inLst = T.toList( options['i'] )
inLst = [ T.absfile( f ) for f in inLst ]
else:
inLst = []
sourceIn = T.absfile( options['s'] )
sourceOut = None
if options.has_key('so'):
sourceOut = T.absfile( options['so'] )
prefix = options['o']
print 'Preparing source ' + str(os.path.basename(sourceIn))\
+ ' -> ' + str(sourceOut)
Default options:
"""
for key, value in o.items():
print "\t-", key, "\t", value
sys.exit(0)
if __name__ == '__main__':
options = T.cmdDict({'o': [os.getcwd()]})
if '?' in options or 'help' in options:
_use(options)
folders = T.toList(options['o'])
if not os.path.exists(folders[0] + '/templates'):
print 'Current directory is not a valid modeling folder.'
_use(options)
T.flushPrint("Creating folders and links...\n")
for f in folders:
sv = VS(outFolder=f)
sv.go(f)
T.flushPrint("done\n")
if __name__ == '__main__':
if len(sys.argv) < 2:
_use()
# options = test()
options = T.cmdDict( defOptions() )
##### COMPLEX LIST ####
options['if'] = int( options['if'] )
options['if_bb'] = int( options['if_bb'] )
## Load complex dictionary
c_lst = T.load( options['i'] )
pkeys = T.toList( options['v'] )
keys = [ 'soln', 'rms', 'rms_if', 'rms_if_bb' ] + pkeys
## get complex data
data = [ transpose(c_lst.valuesOf('soln')) ]
for key in keys[1:]:
data += [ transpose(c_lst.valuesOf(key)) ]
for i in range( len(data) ):
if sum( [ j==None for j in data[i] ] ) > 0:
print '\n################################'
print 'Info data for %s == None for at least one complex'%keys[i]
print 'Plotting will fail for this key.'
print '################################\n'
else:
##############
## MAIN
##############
if __name__ == '__main__':
options = T.cmdDict( {'o':'complexes_pooled.cl',
'mo':'changed_models',
'ldic':'pcr_lig/models.dic',
'rdic':'pcr_rec/models.dic'} )
if len( sys.argv ) < 2:
syntax()
fs = [ T.absfile( f ) for f in T.toList( options['i'] ) ]
result = ComplexList()
rec_dic = T.load( T.absfile( options['rdic'] ) )
lig_dic = T.load( T.absfile( options['ldic'] ) )
for f in fs:
T.flushPrint('Loading %s ...' % f )
cl = T.load( f )
cl = reduceComplexList( cl )
result += cl
for k in [
'debug', 'verbose', 'cast', 'split', 'shuffle', 'shift', 'heavy',
'solvent', 'protein'
]:
if k in options:
options[k] = 1
else:
options[k] = 0
for k in ['chains', 'ex1', 'ex2', 'ex']:
if k in options:
options[k] = t.toIntList(options[k])
if 'atoms' in options:
options['atoms'] = t.toList(options['atoms'])
if 'ex1' in options and 'ex2' in options:
options['ex'] = (options['ex1'], options['ex2'])
else:
options['ex'] = options.get('ex', options.get('ex1', None))
if 'log' in options:
options['log'] = LogFile(options['log'])
f_in = options['i']
del options['i']
a = AmberEntropist(f_in, **options)
a.run()
if start or end or (step != 1):
t = t.takeFrames(range(start, e, step))
return t
############
### MAIN ###
use()
o = T.cmdDict({'o': 'traj_ensemble.dat'})
out = T.absfile(o['o'])
inLst = T.toList(o['i'])
start = int(o.get('s', '0'))
end = o.get('e', None)
if end:
end = int(end)
step = int(o.get('step', 1))
ref = o.get('ref', None)
if ref:
ref = PDBModel(T.absfile(ref))
if 'prot' in o:
ref = ref.compress(ref.maskProtein())
result_xyz = []
result_frameNames = []
result_ref = None
if __name__ == '__main__':
if len(sys.argv) < 2:
_use()
# options = test()
options = T.cmdDict(defOptions())
##### COMPLEX LIST ####
options['if'] = int(options['if'])
options['if_bb'] = int(options['if_bb'])
## Load complex dictionary
c_lst = T.load(options['i'])
pkeys = T.toList(options['v'])
keys = ['soln', 'rms', 'rms_if', 'rms_if_bb'] + pkeys
## get complex data
data = [transpose(c_lst.valuesOf('soln'))]
for key in keys[1:]:
data += [transpose(c_lst.valuesOf(key))]
for i in range(len(data)):
if sum([j == None for j in data[i]]) > 0:
print '\n################################'
print 'Info data for %s == None for at least one complex' % keys[i]
print 'Plotting will fail for this key.'
print '################################\n'
else:
if start or end or (step != 1):
t = t.takeFrames( range( start, e, step ) )
return t
############
### MAIN ###
use()
o = T.cmdDict( {'o':'traj_ensemble.dat'} )
out = T.absfile( o['o'] )
inLst = T.toList( o['i'] )
start = int( o.get('s','0') )
end = o.get( 'e', None )
if end:
end = int( end )
step = int( o.get('step',1) )
ref = o.get('ref',None)
if ref:
ref = PDBModel( T.absfile( ref ) )
if 'prot' in o:
ref = ref.compress( ref.maskProtein() )
result_xyz = []
result_frameNames = []
def cluster( tc, options ):
n_cluster = int( options['n'] )
allowedAtoms = T.toList( options.get('a',[]) )
if allowedAtoms:
mask = tc.traj.ref.mask( lambda a: a['name'] in allowedAtoms )
else:
mask = selectedAtoms( tc.traj.ref )
saveIn = T.absfile( options['o'] ) + '/'
conv = float( options['conv'] )
tc.cluster( n_cluster, aMask=mask, converged=conv )
## collect center frame index for each cluster
frames = [ members[0] for members in tc.memberFrames() ]
result = tc.traj.takeFrames( frames ) ## trajectory of cluster centers
model_dic = {}
dic_index = 1
if options.has_key('ref'):
## use user-provided reference structure
if os.path.isfile( T.absfile(options['ref']) ):
print '\nUsing user specified reference pdb'
m = PDBModel( options['ref'] )
m.remove( m.maskH2O() )
## use reference in trajectory
else:
print '\nUsing reference in trajectory'
m = tc.traj.ref
m = dumpModel( m, options, saveIn+m.getPdbCode()+'_ref.model')
## add ref as first model in dictionary
model_dic[dic_index] = m
dic_index += 1
## save the individual models and add them to the model dictionary
for i in range(0, result.lenFrames() ):
m = result.getPDBModel(i)
m = dumpModel(m, options, saveIn +
T.stripFilename(result.frameNames[i]) +'.model' )
model_dic[dic_index] = m
dic_index += 1
## save model dictionary
fdic = options['dic'] or m.getPdbCode() + '_models.dic'
T.dump( model_dic, T.absfile( fdic ) )
## REDUNDANT CODE AS MULTIDOCK NOW WRITES THE HEX PDB FILES
##
## ## save all models in the dictionary as HEX pdb files
## for k in model_dic.keys():
## m = model_dic[k]
## ## remove hydrogens and sort atoms in standard order
## m.remove( m.maskH() )
## m = molUtils.sortAtomsOfModel(m)
## setChainID( m )
## ## save single hex pdbs
## if options['hex']:
## fhex = options['hex'] + '_%03d' %(k)
## else:
## fhex = m.getPdbCode() + '_%03d_hex.pdb'%(k)
## hexTools.createHexPdb_single( m, T.absfile( fhex ) )
# fhex = options['hex'] or m.getPdbCode() + '_hex.pdb'
# hexTools.createHexPdb( model_dic, T.absfile( fhex ) )
return result
_use()
options = t.cmdDict( defaultOptions() )
## ## current keys used for scoring
## scoreKeys = ['eProsa', 'ePairScore', 'foldX', 'cons_ent', 'cons_max']
## load docking solutions
t.flushPrint( "\nLoading complex list %s ... " % t.absfile( options['i'] ) )
complex_lst = t.load( options['i'] )
t.flushPrint( "done\n" )
## validate and expand list of keys to be calculated
force = []
if options.has_key('f'):
raw_force = t.toList( options['f'] )
## check that the key is valid
validKeys = ['fnac_4.5', 'fnac_10', 'fnrc_4.5',
'fnarc_9', 'fnarc_10', 'c_ratom_9', 'c_ratom_10',
'eProsa', 'ePairScore', 'foldX',
'cons_ent', 'cons_max', 'cons_abs',
'rms_if', 'rms_if_bb']
for key in raw_force:
if key in validKeys:
force += [ key ]
else:
matching = [ v for v in validKeys if key == v[:len(key)] ]
if matching:
force += matching
if len (sys.argv) < 2:
_use( default )
options = T.cmdDict( default )
## where to run the calculation
base_folder = T.absfile( options['r'] )+'/'
## template gly-xxx-gly
template = os.path.abspath( options['i'] )
## label the result dictionaty files
label = '_' + options['l']
## mask to used to delete glycines
mask = [ int(i) for i in T.toList( options['mask'] )]
## create random prptides from template
#randomPeptides( template, base_folder )
## collect average surfaces
MS, AS, MS_sd, AS_sd = randomSurfaces( base_folder, label, mask )
## save dictionary with all 20 amino acids
T.dump( MS, base_folder + 'MS%s.dic'%label)
T.dump( AS, base_folder + 'AS%s.dic'%label )
T.dump( MS_sd, base_folder + 'MS_sd%s.dic'%label )
T.dump( AS_sd, base_folder + 'AS_sd%s.dic'%label )
if k in options:
options[k] = int( options[k] )
for k in ['debug', 'verbose', 'cast', 'split', 'shuffle', 'shift', 'heavy',
'solvent', 'protein']:
if k in options:
options[k] = 1
else:
options[k] = 0
for k in ['chains','ex1', 'ex2', 'ex']:
if k in options:
options[k] = t.toIntList( options[k] )
if 'atoms' in options:
options['atoms'] = t.toList( options['atoms'] )
if 'ex1' in options and 'ex2' in options:
options['ex'] = ( options['ex1'], options['ex2'] )
else:
options['ex'] = options.get( 'ex', options.get('ex1', None) )
if 'log' in options:
options['log'] = LogFile( options['log'] )
f_in = options['i']
del options['i']
a = AmberEntropist( f_in, **options )
a.run()
## MAIN
##############
if __name__ == '__main__':
options = T.cmdDict({
'o': 'complexes_pooled.cl',
'mo': 'changed_models',
'ldic': 'pcr_lig/models.dic',
'rdic': 'pcr_rec/models.dic'
})
if len(sys.argv) < 2:
syntax()
fs = [T.absfile(f) for f in T.toList(options['i'])]
result = ComplexList()
rec_dic = T.load(T.absfile(options['rdic']))
lig_dic = T.load(T.absfile(options['ldic']))
for f in fs:
T.flushPrint('Loading %s ...' % f)
cl = T.load(f)
cl = reduceComplexList(cl)
result += cl
return 1
##########
## MAIN ##
options = tools.cmdDict()
if not 'i' in options:
print """Re-create LocalPath in PDBModel or Complex from the original
file name (and the current environment variables).
relocalize.py -i |file1 file2 ..|
"""
sys.exit(0)
files = tools.toList(options['i'])
repl = []
if 'r' in options:
relacements = tools.toList(options['r'])
repl = [r.split(':') for r in relacements]
for f in files:
try:
print "Re-localizing ", f
f = tools.absfile(f)
o = tools.load(f)
result = 0
if o.__class__ in [PCRModel, PDBModel]:
result = localizeModel(o, repl, f)
def cluster(tc, options):
n_cluster = int(options['n'])
allowedAtoms = T.toList(options.get('a', []))
if allowedAtoms:
mask = tc.traj.ref.mask(lambda a: a['name'] in allowedAtoms)
else:
mask = selectedAtoms(tc.traj.ref)
saveIn = T.absfile(options['o']) + '/'
conv = float(options['conv'])
tc.cluster(n_cluster, aMask=mask, converged=conv)
## collect center frame index for each cluster
frames = [members[0] for members in tc.memberFrames()]
result = tc.traj.takeFrames(frames) ## trajectory of cluster centers
model_dic = {}
dic_index = 1
if options.has_key('ref'):
## use user-provided reference structure
if os.path.isfile(T.absfile(options['ref'])):
print '\nUsing user specified reference pdb'
m = PDBModel(options['ref'])
m.remove(m.maskH2O())
## use reference in trajectory
else:
print '\nUsing reference in trajectory'
m = tc.traj.ref
m = dumpModel(m, options, saveIn + m.getPdbCode() + '_ref.model')
## add ref as first model in dictionary
model_dic[dic_index] = m
dic_index += 1
## save the individual models and add them to the model dictionary
for i in range(0, result.lenFrames()):
m = result.getPDBModel(i)
m = dumpModel(
m, options,
saveIn + T.stripFilename(result.frameNames[i]) + '.model')
model_dic[dic_index] = m
dic_index += 1
## save model dictionary
fdic = options['dic'] or m.getPdbCode() + '_models.dic'
T.dump(model_dic, T.absfile(fdic))
## REDUNDANT CODE AS MULTIDOCK NOW WRITES THE HEX PDB FILES
##
## ## save all models in the dictionary as HEX pdb files
## for k in model_dic.keys():
## m = model_dic[k]
## ## remove hydrogens and sort atoms in standard order
## m.remove( m.maskH() )
## m = molUtils.sortAtomsOfModel(m)
## setChainID( m )
## ## save single hex pdbs
## if options['hex']:
## fhex = options['hex'] + '_%03d' %(k)
## else:
## fhex = m.getPdbCode() + '_%03d_hex.pdb'%(k)
## hexTools.createHexPdb_single( m, T.absfile( fhex ) )
# fhex = options['hex'] or m.getPdbCode() + '_hex.pdb'
# hexTools.createHexPdb( model_dic, T.absfile( fhex ) )
return result
if importas == module:
importas = ''
else:
importas = ' as ' + importas
return l.replace('from ' + module + ' import *',
'import ' + module + importas)
options = t.cmdDict({'m': 'Numeric', 'as': 'N', 'e': ['Complex']})
module = options['m']
importas = options.get('as', module)
srcfiles = t.toList(options.get('i', None))
exclude = t.toList(options['e'])
try:
exec('import ' + module)
for fname in srcfiles:
fname = t.absfile(fname)
shutil.copy(fname, fname + '_')
methods = re_lst(module, exclude)
fold = open(fname + '_')
fnew = open(fname, 'w')
_use()
options = t.cmdDict(defaultOptions())
## ## current keys used for scoring
## scoreKeys = ['eProsa', 'ePairScore', 'foldX', 'cons_ent', 'cons_max']
## load docking solutions
t.flushPrint("\nLoading complex list %s ... " % t.absfile(options['i']))
complex_lst = t.load(options['i'])
t.flushPrint("done\n")
## validate and expand list of keys to be calculated
force = []
if options.has_key('f'):
raw_force = t.toList(options['f'])
## check that the key is valid
validKeys = [
'fnac_4.5', 'fnac_10', 'fnrc_4.5', 'fnarc_9', 'fnarc_10', 'c_ratom_9',
'c_ratom_10', 'eProsa', 'ePairScore', 'foldX', 'cons_ent', 'cons_max',
'cons_abs', 'rms_if', 'rms_if_bb'
]
for key in raw_force:
if key in validKeys:
force += [key]
else:
matching = [v for v in validKeys if key == v[:len(key)]]
if matching:
force += matching
sort = options.has_key('s')
outFolder = None
if options.has_key('o'):
outFolder = T.absfile( options['o'] )
## get all PDBs and models directly from a directory (avoids shell limits)
if 'd' in options:
d = T.absfile( options['d'] )
files = os.listdir( d )
files = [ x for x in l if x[-4:].upper() == '.PDB' or \
x[-6:] == '.MODEL' or x[-7:].upper() == '.PDB.GZ' ]
files = [ os.path.join( d, f ) for f in files ]
else:
## extract file names
files = T.toList( options['i'] )
## add needed amount of hosts
add_hosts = options.has_key('a')
skipWat = options.has_key('wat')
fileDic = {}
for f in files:
fileDic[T.absfile( f )] = ''
print "Initialize Job queue.."
master = StructMaster(fileDic, chunk , hosts.cpus_all[ : hostNumber ] ,
outFolder, skipWat=skipWat, amber=amber,
sort=sort, show_output=windows, add_hosts=add_hosts )
if '?' in options or 'help' in options:
_use( options )
if not os.path.exists(f + VS.F_RESULT_FOLDER) and not options.has_key('d'):
print 'Current directory is not a valid modeling project folder.'
_use( options )
## Try to add project folders
## look for default cross-validation projects
d = []
if os.path.exists( f + VS.F_RESULT_FOLDER ):
d = [ f ]
if options.has_key('d'):
folders = T.toList(options['d'])
else:
folders = d
T.flushPrint("Starting job...\n")
for f in folders:
a = A(outFolder=f)
a.go()
T.flushPrint("Done.\n")
## show result in PyMol
if options.has_key('s'):
p=Pymoler()
sort = options.has_key('s')
outFolder = None
if options.has_key('o'):
outFolder = T.absfile(options['o'])
## get all PDBs and models directly from a directory (avoids shell limits)
if 'd' in options:
d = T.absfile(options['d'])
files = os.listdir(d)
files = [ x for x in l if x[-4:].upper() == '.PDB' or \
x[-6:] == '.MODEL' or x[-7:].upper() == '.PDB.GZ' ]
files = [os.path.join(d, f) for f in files]
else:
## extract file names
files = T.toList(options['i'])
## add needed amount of hosts
add_hosts = options.has_key('a')
skipWat = options.has_key('wat')
fileDic = {}
for f in files:
fileDic[T.absfile(f)] = ''
print "Initialize Job queue.."
master = StructMaster(fileDic,
chunk,
hosts.cpus_all[:hostNumber],
