def makePPI(self): ### Generates files for Human-HIV PPI analysis
'''Generates files for Human-HIV PPI analysis.'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
seqlist = rje_seq.SeqList(self.log,self.cmd_list+['seqin=%s' % self.getStr('HIVSeq'),'autoload=T'])
if not seqlist.seqs(): return False
seqmap = seqlist.seqNameDic('Max')
mdb = self.db('HHPIDMap')
### ~ [2] Process ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for hivacc in mdb.index('AccHIV'):
# map HIV accession numbers on to sequences seqNameDic
accnum = string.split(hivacc,'.')[0]
hivseq = seqmap[accnum]
# extract short HIV name from sequence ID
hivgene = string.split(hivseq.shortName(),'_')[0].upper()
# create directory named after HIV gene
#self.progLog('\r#PPI','Generating human-HIV PPI fasta files for %s' % (hivgene))
rje.mkDir(self,'%s/' % hivgene,log=True)
# copy human PPI files into directories, adding HIV gene
ex = 0.0; etot = len(mdb.index('AccHIV')[hivacc])
for entry in mdb.indexEntries('AccHIV',hivacc):
self.progLog('\r#PPI','Generating human-HIV PPI fasta files for %s %s PPI' % (rje.iStr(etot),hivgene))
pfile = self.getStr('PPIDir') + entry['Symbol'] + '.ppi.fas'
if rje.exists(pfile):
FAS = open('%s/%s.%s.ppi.fas' % (hivgene,hivgene.lower(),entry['Symbol']),'w')
FAS.write('>%s\n%s\n' % (hivseq.info['Name'],hivseq.getSequence()))
FAS.write(open(pfile,'r').read())
FAS.close()
else: self.errorLog('Cannot find human PPI file for %s interactor "%s"' % (entry['HIV'],entry['Symbol']),printerror=False)
self.printLog('\r#PPI','Generated human-HIV PPI fasta files for %s %s (%s) PPI.' % (rje.iStr(etot),entry['HIV'],hivgene))
except: self.errorLog('%s.makePPI error' % self); return False
def save(self): ### Saves parsed REST output to files
'''Saves parsed REST output to files.'''
rbase = '%s%s' % (self.getStr('RestOutDir'),
rje.baseFile(self.getStr('RestBase'),
strip_path=True,
keepext=True))
rje.mkDir(self, self.getStr('RestOutDir'))
outputs = rje.sortKeys(self.dict['Output'])
if self.getStrLC('Rest') in outputs: outputs = [self.getStrLC('Rest')]
elif self.getStrLC('Rest') in ['full', 'text']:
outfile = '%s.rest' % rbase
open(outfile, 'w').write(self.restFullOutput())
self.printLog('#OUT', '%s: %s' % (self.getStrLC('Rest'), outfile))
return True
elif self.getStrLC('Rest'):
self.printLog(
'#OUTFMT', 'REST output format "%s" not recognised.' %
self.getStrLC('Rest'))
if self.i() < 0 or not rje.yesNo('Output all parsed outputs?'):
return False
outfile = '%s.rest' % rbase
open(outfile, 'w').write(self.restFullOutput())
self.printLog('#OUT', 'full: %s' % (outfile))
return True
for rkey in outputs:
if rkey in self.dict['Outfile']:
rje.backup(self, self.dict['Outfile'][rkey])
open(self.dict['Outfile'][rkey],
'w').write(self.dict['Output'][rkey])
self.printLog('#OUT',
'%s: %s' % (rkey, self.dict['Outfile'][rkey]))
elif rkey not in ['intro']:
self.warnLog('No outfile parsed/generated for %s output' %
rkey)
def ppi(self): ### Remaining protein-protein interactions
'''Remaining protein-protein interactions.'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if not self.dict['PPI']: return
outdir = 'SLiMPID_PPI'
rje.mkDir(self, outdir)
badname = []
### ~ [2] Process ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for hub in rje.sortKeys(self.dict['PPI']):
gene = self.dict['Gene'][hub]
acc = []
for name in self.dict['PPI'][hub]:
if not name: continue
if name in self.dict['Seq']:
acc.append(self.dict['Seq'][name].info['AccNum'])
elif name not in badname:
badname.append(name)
open('%s/%s.ppi.acc' % (outdir, gene),
'w').write(string.join(acc, '\n'))
self.printLog(
'#PPI',
'%s => %d individual interactors' % (gene, len(acc)))
if badname:
badname.sort()
self.printLog(
'#BAD', '%d "bad" protein names: %s' %
(len(badname), string.join(badname, '; ')))
except:
self.errorLog('Problem with SLiMPID.setup()', quitchoice=True)
def complexFasta(
self): ### Outputs parsed complex datasets in Fasta format
'''Outputs parsed complex datasets in Fasta format.'''
try:
### Setup ###
datpath = self.info['OutDir'] + rje.makePath('HPRD_Complexes/')
rje.mkDir(self, datpath)
### Output PPI Datasets ###
for complex in rje.sortKeys(self.dict['Complex']):
mylist = []
for p2 in self.dict['Complex'][complex]:
if self.opt['AllIso']:
mylist += self.dict['HPRD'][p2]['Seq']
else:
mylist.append(self.dict['HPRD'][p2]['Seq'])
sfile = '%s%s_hprd.fas' % (datpath, complex)
if mylist:
self.obj['SeqList'].saveFasta(seqs=mylist, seqfile=sfile)
self.log.printLog('#FAS', 'HPRD complex fasta output complete.')
except:
self.log.errorLog('Error in HPRD.complexFasta()',
printerror=True,
quitchoice=False)
raise
def saveFasta(self): ### Outputs parsed PPI datasets in Fasta format
'''Outputs parsed PPI datasets in Fasta format.'''
try:
### Setup ###
datpath = self.info['OutDir'] + rje.makePath('HPRD_Datasets/')
rje.mkDir(self, datpath)
## Check Seqs ##
for p1 in rje.sortKeys(self.dict['PPI']):
if 'Seq' not in self.dict['HPRD'][p1]: #!# KeyError #!#
print p1, self.dict['HPRD'][p1]
self.deBug('No Seq for %s' % p1)
### All sequences ###
self.obj['SeqList'].saveFasta()
### Output PPI Datasets ###
for p1 in rje.sortKeys(self.dict['PPI']):
mylist = []
for p2 in self.dict['PPI'][p1]:
if self.opt['AllIso']:
mylist += self.dict['HPRD'][p2]['Seq']
else:
mylist.append(self.dict['HPRD'][p2]['Seq'])
sfile = '%s%s_hprd.fas' % (datpath,
self.dict['HPRD'][p1]['gene'])
if mylist:
self.obj['SeqList'].saveFasta(seqs=mylist, seqfile=sfile)
self.log.printLog('#FAS', 'HPRD PPI fasta output complete.')
except:
self.log.errorLog('Error in HPRD.saveFasta()',
printerror=True,
quitchoice=False)
def makePPIDatasets(self): ### Generate PPI datasets from pairwise data
'''Generate PPI datasets from pairwise data.'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
rje.mkDir(self, 'YeastPPI/')
seqdict = self.dict['SeqDict']
### ~ [2] Parse data ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
(hx, htot, fx) = (0.0, len(self.dict['PPI']), 0)
for hub in rje.sortKeys(self.dict['PPI']):
self.progLog(
'\r#FAS', 'Generating %s PPI fasta files: %.2f' %
(rje.integerString(fx), hx / htot))
hx += 100.0
if len(self.dict['PPI'][hub]) < 3: continue
seqs = []
for spoke in self.dict['PPI'][hub]:
if spoke not in seqdict: continue
seqs.append(seqdict[spoke])
if len(seqs) < 3: continue
self.obj['SeqList'].saveFasta(seqs,
rje.makePath('YeastPPI/%s.fas' %
hub,
wholepath=True),
log=False)
fx += 1
self.printLog(
'\r#FAS',
'Generation of %s PPI fasta files from %s hubs complete.' %
(rje.integerString(fx), rje.integerString(htot)))
except:
self.errorLog(rje_zen.Zen().wisdom())
raise # Delete this if method error not terrible
def dpi(self): ### Domain-protein interactions
'''Domain-protein interactions.'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if not self.dict['Domain']: return
outdir = 'SLiMPID_DPI'
rje.mkDir(self, outdir)
dpi = {} # Dictionary of {domain:[interactors]}
badname = []
### ~ [2] Process ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for dom in rje.sortKeys(self.dict['Domain']):
dpi[dom] = []
for hub in self.dict['Domain'][dom]:
if hub in self.dict['PPI']:
dpi[dom] += self.dict['PPI'][
hub] # Add with redundancy
for spoke in dpi[dom][0:]:
if dpi[dom].count(spoke) == 1:
dpi[dom].remove(
spoke) # Must have 2+ domain interactions
for hub in self.dict['Domain'][dom]:
if hub not in self.dict['PPI']: continue
for spoke in self.dict['PPI'][hub][0:]:
if spoke in dpi[dom]:
self.dict['PPI'][hub].remove(spoke)
if spoke in self.dict['PPI'] and hub in self.dict[
'PPI'][spoke]:
self.dict['PPI'][spoke].remove(hub)
dpi[dom] = rje.sortUnique(dpi[dom], False, False)
acc = []
for name in dpi[dom]:
if not name: continue
if name in self.dict['Seq']:
acc.append(self.dict['Seq'][name].info['AccNum'])
elif name not in badname:
badname.append(name)
open('%s/%s.dpi.acc' % (outdir, dom),
'w').write(string.join(acc, '\n'))
self.printLog('#DPI',
'%s domain => %d interactors' % (dom, len(acc)))
if badname:
badname.sort()
self.printLog(
'#BAD', '%d "bad" protein names: %s' %
(len(badname), string.join(badname, '; ')))
### ~ [3] Cleanup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
hx = len(self.dict['PPI'])
for hub in rje.sortKeys(self.dict['PPI']):
if hub and self.dict['PPI'][hub]: continue
self.dict['PPI'].pop(hub)
self.printLog('#DPI',
'No %s PPI left after DPI removed' % hub,
screen=False)
self.printLog(
'#PPX', '%s of %s PPI hubs remain after DPI removed' %
(rje.integerString(len(
self.dict['PPI'])), rje.integerString(hx)))
except:
self.errorLog('Problem with SLiMPID.dpi()', quitchoice=True)
def setup(self): ### Main class setup method.
'''Main class setup method.'''
try:### ~ [1] Setup SeqList ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
self.obj['SeqList'] = rje_seq.SeqList(self.log,['keepblast=T']+self.cmd_list+['autofilter=F','align=F','haqbat=None'])
self.obj['SeqList']._checkForDup(True)
if not self.seqNum(): self.errorLog('No sequences loaded!',printerror=False); return False
if self.opt['AddQueries'] and self.name() not in self.obj['SeqList'].list['Blast2Fas']: self.obj['SeqList'].list['Blast2Fas'].append(self.name())
### ~ [2] Setup Results Directory ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if self.info['HaqDir'].lower() in ['','none']: self.info['HaqDir'] = '%s_HAQESAC/' % rje.baseFile(self.name(), strip_path=True)
rje.mkDir(self,self.info['HaqDir'])
return True # Setup successful
except: self.errorLog('Problem during %s setup.' % self); return False # Setup failed
def saveReadMe(self,
filename='pydocs.txt',
append=False): ### Prints docs for modules to file
'''
Prints docs for modules to file.
>> filename:str = output file name
>> append:boolean
'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
pydoc = self.obj['PyDoc']
if append:
self.printLog('#DOC', 'Appending docstrings to %s' % filename)
PYDOC = open(filename, 'a')
else:
rje.mkDir(self, filename)
self.printLog('#DOC', 'Writing docstrings to %s' % filename)
PYDOC = open(filename, 'w')
PYDOC.write(self.readMeHeader())
db = self.db('Module')
dx = 0
### ~ [2] Output Docstrings ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for sourcedir in pydoc.list['SourceDir']:
PYDOC.write('-%s:\n\n' % sourcedir)
for pyfile in db.dataKeys():
entry = db.data(pyfile)
module = entry['Module']
if not pyfile.find(sourcedir) >= 0 or not os.path.exists(
'%s%s%s.py' % (pydoc.getStr('PyPath'),
rje.makePath(sourcedir), module)):
continue
## ~ [2a] ~ Module docstring ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
mtxt = '### ~~~ Module %s ~ [%s] ~~~ ###' % (module,
pyfile)
while len(mtxt) < 122:
mtxt = mtxt[:5] + '~' + mtxt[5:-5] + '~' + mtxt[-5:]
try:
PYDOC.write('%s\n\n%s\n' % (mtxt, entry['DocString']))
dx += 1
except:
self.errorLog('Cannot write DocString for %s' % module,
printerror=False)
PYDOC.write('%s\n\nDocString Error!\n' % (mtxt))
dx += 1
PYDOC.write('\n\n\n')
PYDOC.close()
self.printLog(
'#DOC', 'Output to %s complete: %s modules.' %
(filename, rje.iStr(dx)))
except:
self.errorLog('Error in %s.saveDocs()' % self.prog())
def setupSourceData(self): ### Main class setup method. #V0.0
'''Setup and optionally download source data.'''
try:### ~ [0] Setup Source Data ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
db = self.obj['DB']
sdb = db.addEmptyTable('Source',['Name','File','Status','Entries','URL'],keys=['Name'],log=False) # Store Source info
rje.mkDir(self,self.getStr('SourcePath'),True)
self.printLog('#~~#','## ~~~~~~~~~~~~~~~~~~~~~~ SETUP TAXONOMIC DATA ~~~~~~~~~~~~~~~~~~~~~ ##',timeout=False)
### ~ [1] Uniprot species codes file ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if not self.sourceDataFile('SpecFile'): raise IOError
### ~ [2] NCBI TaxID mapping file ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if not self.sourceDataFile('TaxMap'): raise IOError
if not self.sourceDataFile('NameMap'): raise IOError
return True
except: self.errorLog('%s.setupSourceData failure' % self); return False
def makePPI(self): ### Generates files for Human-HIV PPI analysis
'''Generates files for Human-HIV PPI analysis.'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
seqlist = rje_seq.SeqList(
self.log, self.cmd_list +
['seqin=%s' % self.getStr('HIVSeq'), 'autoload=T'])
if not seqlist.seqs(): return False
seqmap = seqlist.seqNameDic('Max')
mdb = self.db('HHPIDMap')
### ~ [2] Process ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for hivacc in mdb.index('AccHIV'):
# map HIV accession numbers on to sequences seqNameDic
accnum = string.split(hivacc, '.')[0]
hivseq = seqmap[accnum]
# extract short HIV name from sequence ID
hivgene = string.split(hivseq.shortName(), '_')[0].upper()
# create directory named after HIV gene
#self.progLog('\r#PPI','Generating human-HIV PPI fasta files for %s' % (hivgene))
rje.mkDir(self, '%s/' % hivgene, log=True)
# copy human PPI files into directories, adding HIV gene
ex = 0.0
etot = len(mdb.index('AccHIV')[hivacc])
for entry in mdb.indexEntries('AccHIV', hivacc):
self.progLog(
'\r#PPI',
'Generating human-HIV PPI fasta files for %s %s PPI' %
(rje.iStr(etot), hivgene))
pfile = self.getStr(
'PPIDir') + entry['Symbol'] + '.ppi.fas'
if rje.exists(pfile):
FAS = open(
'%s/%s.%s.ppi.fas' %
(hivgene, hivgene.lower(), entry['Symbol']), 'w')
FAS.write('>%s\n%s\n' %
(hivseq.info['Name'], hivseq.getSequence()))
FAS.write(open(pfile, 'r').read())
FAS.close()
else:
self.errorLog(
'Cannot find human PPI file for %s interactor "%s"'
% (entry['HIV'], entry['Symbol']),
printerror=False)
self.printLog(
'\r#PPI',
'Generated human-HIV PPI fasta files for %s %s (%s) PPI.' %
(rje.iStr(etot), entry['HIV'], hivgene))
except:
self.errorLog('%s.makePPI error' % self)
return False
def fpi(self): ### Family-protein interactions
'''Family-protein interactions.'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if not self.dict['Domain']: return
outdir = 'SLiMPID_FPI'
rje.mkDir(self,outdir)
fpi = {} # Dictionary of {family:[interactors]}
badname = []
### ~ [2] Process ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for qry in rje.sortKeys(self.dict['PPI']):
try:
fam = self.dict['Fam'][qry]
if len(fam) < 2: continue
except: self.errorLog('Problem with "%s" protein family' % qry); continue
fpi[qry] = []
for hub in fam:
if hub not in self.dict['PPI']: continue
fpi[qry] += self.dict['PPI'][hub] # Add with redundancy
for spoke in fpi[qry][0:]:
if fpi[qry].count(spoke) == 1: fpi[qry].remove(spoke) # Must have 2+ family interactions
for hub in fam:
if hub not in self.dict['PPI']: continue
for spoke in self.dict['PPI'][hub][0:]:
if spoke in fpi[qry]:
self.dict['PPI'][hub].remove(spoke)
if spoke in self.dict['PPI'] and hub in self.dict['PPI'][spoke]: self.dict['PPI'][spoke].remove(hub)
fpi[qry] = rje.sortUnique(fpi[qry],False,False)
acc = []
gene = self.dict['Gene'][qry]
for name in fpi[qry]:
if not name: continue
if name in self.dict['Seq']: acc.append(self.dict['Seq'][name].info['AccNum'])
elif name not in badname: badname.append(name)
open('%s/%s.fpi.acc' % (outdir,gene),'w').write(string.join(acc,'\n'))
self.printLog('#FPI','%s family => %d interactors' % (gene,len(acc)))
if badname:
badname.sort()
self.printLog('#BAD','%d "bad" protein names: %s' % (len(badname),string.join(badname,'; ')))
### ~ [3] Cleanup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
hx = len(self.dict['PPI'])
for hub in rje.sortKeys(self.dict['PPI']):
if hub and self.dict['PPI'][hub]: continue
self.dict['PPI'].pop(hub)
self.printLog('#FPI','No %s PPI left after FPI removed' % hub)
self.printLog('#PPX','%s of %s PPI hubs remain after FPI removed' % (rje.integerString(len(self.dict['PPI'])),rje.integerString(hx)))
except: self.errorLog('Problem with SLiMPID.fpi()',quitchoice=True)
def domainFasta(self): ### Outputs parsed domain and domain PPI datasets in Fasta format
'''Outputs parsed PPI datasets in Fasta format.'''
try:
### ~ Tab delimited domain-HPRD pairs ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
headers = ['Domain','HPRD','Gene']
dfile = self.info['OutDir'] + 'HPRD.domains.tdt'
rje.delimitedFileOutput(self,dfile,headers,'\t')
sfile = self.info['OutDir'] + 'HPRD.domsource.tdt'
shead = ['Domain','Source']
rje.delimitedFileOutput(self,sfile,shead,'\t')
dx = 0.0
for domain in rje.sortKeys(self.dict['Domains']):
self.log.printLog('\r#DOM','HPRD Domain output (%s): %.1f%%' % (dfile,dx/len(self.dict['Domains'])),newline=False,log=False)
dx += 100.0
for hid in self.dict['Domains'][domain]:
datadict = {'Domain':domain,'HPRD':hid,'Gene':self.dict['HPRD'][hid]['gene']}
rje.delimitedFileOutput(self,dfile,headers,'\t',datadict)
for source in self.dict['DomainSource'][domain]:
datadict = {'Domain':domain,'Source':source}
rje.delimitedFileOutput(self,sfile,shead,'\t',datadict)
self.log.printLog('\r#DOM','HPRD Domain output (%s): %s domains.' % (dfile,rje.integerString(len(self.dict['Domains']))))
### ~ Domain PPI Dataset Outputs ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
datpath = self.info['OutDir'] + rje.makePath('HPRD_Domain_Datasets/')
rje.mkDir(self,datpath)
for domain in rje.sortKeys(self.dict['Domains']):
## Generate a list of all interactors with domain-containing proteins ##
plist = []
for p1 in self.dict['Domains'][domain]:
if p1 not in self.dict['PPI']: continue
for p2 in self.dict['PPI'][p1]:
if p2 not in plist: plist.append(p2)
plist.sort()
## Generate Sequence list and output ##
mylist = []
for p in plist:
if self.opt['AllIso']: mylist += self.dict['HPRD'][p]['Seq']
else: mylist.append(self.dict['HPRD'][p]['Seq'])
sfile = '%s%s_hprd.fas' % (datpath,domain)
if mylist: self.obj['SeqList'].saveFasta(seqs=mylist,seqfile=sfile)
else: self.log.printLog('#DOM','No PPI partners for domain "%s"' % domain)
self.log.printLog('\r#DOM','HPRD Domain fasta output complete.')
except:
self.log.errorLog('Error in HPRD.saveFasta()',printerror=True,quitchoice=False)
raise
def makePPIDatasets(self): ### Generate PPI datasets from pairwise data
'''Generate PPI datasets from pairwise data.'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
rje.mkDir(self,'YeastPPI/')
seqdict = self.dict['SeqDict']
### ~ [2] Parse data ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
(hx,htot,fx) = (0.0,len(self.dict['PPI']),0)
for hub in rje.sortKeys(self.dict['PPI']):
self.progLog('\r#FAS','Generating %s PPI fasta files: %.2f' % (rje.integerString(fx),hx/htot)); hx += 100.0
if len(self.dict['PPI'][hub]) < 3: continue
seqs = []
for spoke in self.dict['PPI'][hub]:
if spoke not in seqdict: continue
seqs.append(seqdict[spoke])
if len(seqs) < 3: continue
self.obj['SeqList'].saveFasta(seqs,rje.makePath('YeastPPI/%s.fas' % hub,wholepath=True),log=False); fx+=1
self.printLog('\r#FAS','Generation of %s PPI fasta files from %s hubs complete.' % (rje.integerString(fx),rje.integerString(htot)))
except: self.errorLog(rje_zen.Zen().wisdom()); raise # Delete this if method error not terrible
def complexFasta(self): ### Outputs parsed complex datasets in Fasta format
'''Outputs parsed complex datasets in Fasta format.'''
try:
### Setup ###
datpath = self.info['OutDir'] + rje.makePath('HPRD_Complexes/')
rje.mkDir(self,datpath)
### Output PPI Datasets ###
for complex in rje.sortKeys(self.dict['Complex']):
mylist = []
for p2 in self.dict['Complex'][complex]:
if self.opt['AllIso']: mylist += self.dict['HPRD'][p2]['Seq']
else: mylist.append(self.dict['HPRD'][p2]['Seq'])
sfile = '%s%s_hprd.fas' % (datpath,complex)
if mylist: self.obj['SeqList'].saveFasta(seqs=mylist,seqfile=sfile)
self.log.printLog('#FAS','HPRD complex fasta output complete.')
except:
self.log.errorLog('Error in HPRD.complexFasta()',printerror=True,quitchoice=False)
raise
def ppi(self): ### Remaining protein-protein interactions
'''Remaining protein-protein interactions.'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if not self.dict['PPI']: return
outdir = 'SLiMPID_PPI'
rje.mkDir(self,outdir)
badname = []
### ~ [2] Process ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for hub in rje.sortKeys(self.dict['PPI']):
gene = self.dict['Gene'][hub]
acc = []
for name in self.dict['PPI'][hub]:
if not name: continue
if name in self.dict['Seq']: acc.append(self.dict['Seq'][name].info['AccNum'])
elif name not in badname: badname.append(name)
open('%s/%s.ppi.acc' % (outdir,gene),'w').write(string.join(acc,'\n'))
self.printLog('#PPI','%s => %d individual interactors' % (gene,len(acc)))
if badname:
badname.sort()
self.printLog('#BAD','%d "bad" protein names: %s' % (len(badname),string.join(badname,'; ')))
except: self.errorLog('Problem with SLiMPID.setup()',quitchoice=True)
def dpi(self): ### Domain-protein interactions
'''Domain-protein interactions.'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if not self.dict['Domain']: return
outdir = 'SLiMPID_DPI'
rje.mkDir(self,outdir)
dpi = {} # Dictionary of {domain:[interactors]}
badname = []
### ~ [2] Process ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for dom in rje.sortKeys(self.dict['Domain']):
dpi[dom] = []
for hub in self.dict['Domain'][dom]:
if hub in self.dict['PPI']: dpi[dom] += self.dict['PPI'][hub] # Add with redundancy
for spoke in dpi[dom][0:]:
if dpi[dom].count(spoke) == 1: dpi[dom].remove(spoke) # Must have 2+ domain interactions
for hub in self.dict['Domain'][dom]:
if hub not in self.dict['PPI']: continue
for spoke in self.dict['PPI'][hub][0:]:
if spoke in dpi[dom]:
self.dict['PPI'][hub].remove(spoke)
if spoke in self.dict['PPI'] and hub in self.dict['PPI'][spoke]: self.dict['PPI'][spoke].remove(hub)
dpi[dom] = rje.sortUnique(dpi[dom],False,False)
acc = []
for name in dpi[dom]:
if not name: continue
if name in self.dict['Seq']: acc.append(self.dict['Seq'][name].info['AccNum'])
elif name not in badname: badname.append(name)
open('%s/%s.dpi.acc' % (outdir,dom),'w').write(string.join(acc,'\n'))
self.printLog('#DPI','%s domain => %d interactors' % (dom,len(acc)))
if badname:
badname.sort()
self.printLog('#BAD','%d "bad" protein names: %s' % (len(badname),string.join(badname,'; ')))
### ~ [3] Cleanup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
hx = len(self.dict['PPI'])
for hub in rje.sortKeys(self.dict['PPI']):
if hub and self.dict['PPI'][hub]: continue
self.dict['PPI'].pop(hub)
self.printLog('#DPI','No %s PPI left after DPI removed' % hub,screen=False)
self.printLog('#PPX','%s of %s PPI hubs remain after DPI removed' % (rje.integerString(len(self.dict['PPI'])),rje.integerString(hx)))
except: self.errorLog('Problem with SLiMPID.dpi()',quitchoice=True)
def setup(self): ### Main class setup method.
'''Main class setup method.'''
try: ### ~ [1] Setup SeqList ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
self.obj['SeqList'] = rje_seq.SeqList(
self.log, ['keepblast=T'] + self.cmd_list +
['autofilter=F', 'align=F', 'haqbat=None'])
self.obj['SeqList']._checkForDup(True)
if not self.seqNum():
self.errorLog('No sequences loaded!', printerror=False)
return False
if self.opt['AddQueries'] and self.name(
) not in self.obj['SeqList'].list['Blast2Fas']:
self.obj['SeqList'].list['Blast2Fas'].append(self.name())
### ~ [2] Setup Results Directory ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if self.info['HaqDir'].lower() in ['', 'none']:
self.info['HaqDir'] = '%s_HAQESAC/' % rje.baseFile(
self.name(), strip_path=True)
rje.mkDir(self, self.info['HaqDir'])
return True # Setup successful
except:
self.errorLog('Problem during %s setup.' % self)
return False # Setup failed
def gopher(self): ### Sets up data for GOPHER run
'''Sets up data for GOPHER run.'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
rje.mkDir(self,'BLAST/')
rje_blast.BLASTRun(self.log,self.cmd_list).formatDB(fasfile='%s.ygob.fas' % self.info['Basefile'],protein=True,force=False)
rje_blast.BLASTRun(self.log,self.cmd_list).formatDB(fasfile='%s.yeast.fas' % self.info['Basefile'],protein=True,force=False)
seqdict = self.obj['SeqList'].seqNameDic('AccNum')
ymap = self.dict['PillarMap'] = {}
### ~ [2] Convert Pillars to BLAST IDs ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
(px,ptot) = (0.0,len(self.list['Pillars'])); ox = 0
for pillar in self.list['Pillars']:
self.progLog('\r#YGOB','Converting YGOB Pillars for GOPHER: %.2f%%' % (px/ptot)); px += 100
newpillar = []
for yid in pillar:
seq = rje_sequence.Sequence(self.log,self.cmd_list)
seq.opt['Yeast'] = True
#self.deBug(yid)
seq.info['Name'] = yid
seq.extractDetails(gnspacc=True)
#self.deBug(seq.info)
ygob = seq.info['AccNum']
if ygob in self.dict['Rename']: acc = self.dict['Rename'][ygob]
else: acc = ygob
ymap[yid] = acc
if acc not in seqdict: self.printLog('\r#GENE','Non-coding gene %s (%s)? Cannot find in fasta file' % (acc,yid)); continue
try:
newpillar.append(seqdict[acc].shortName())
except:
print yid, ygob, acc
self.errorLog(rje_zen.Zen().wisdom())
if not newpillar: continue
for ygob in pillar:
acc = ymap[ygob]
if acc not in seqdict: continue
if acc in self.list['YeastSeq'] or (not self.list['YeastSeq'] and seqdict[acc].info['SpecCode'] == 'YEAST'):
open(rje.makePath('BLAST/%s.blast.id' % acc,wholepath=True),'w').write(string.join(newpillar,'\n'))
ox += 1
self.progLog('\r#YGOB','Converted YGOB Pillars for GOPHER: %s BLAST ID files.' % rje.iStr(ox))
except: self.errorLog(rje_zen.Zen().wisdom()); raise # Delete this if method error not terrible
def saveFasta(self): ### Outputs parsed PPI datasets in Fasta format
'''Outputs parsed PPI datasets in Fasta format.'''
try:
### Setup ###
datpath = self.info['OutDir'] + rje.makePath('HPRD_Datasets/')
rje.mkDir(self,datpath)
## Check Seqs ##
for p1 in rje.sortKeys(self.dict['PPI']):
if 'Seq' not in self.dict['HPRD'][p1]: #!# KeyError #!#
print p1, self.dict['HPRD'][p1]
self.deBug('No Seq for %s' % p1)
### All sequences ###
self.obj['SeqList'].saveFasta()
### Output PPI Datasets ###
for p1 in rje.sortKeys(self.dict['PPI']):
mylist = []
for p2 in self.dict['PPI'][p1]:
if self.opt['AllIso']: mylist += self.dict['HPRD'][p2]['Seq']
else: mylist.append(self.dict['HPRD'][p2]['Seq'])
sfile = '%s%s_hprd.fas' % (datpath,self.dict['HPRD'][p1]['gene'])
if mylist: self.obj['SeqList'].saveFasta(seqs=mylist,seqfile=sfile)
self.log.printLog('#FAS','HPRD PPI fasta output complete.')
except: self.log.errorLog('Error in HPRD.saveFasta()',printerror=True,quitchoice=False)
def save(self): ### Saves parsed REST output to files
'''Saves parsed REST output to files.'''
rbase = '%s%s' % (self.getStr('RestOutDir'),rje.baseFile(self.getStr('RestBase'),strip_path=True,keepext=True))
rje.mkDir(self,self.getStr('RestOutDir'))
outputs = rje.sortKeys(self.dict['Output'])
if self.getStrLC('Rest') in outputs: outputs = [self.getStrLC('Rest')]
elif self.getStrLC('Rest') in ['full','text']:
outfile = '%s.rest' % rbase
open(outfile,'w').write(self.restFullOutput())
self.printLog('#OUT','%s: %s' % (self.getStrLC('Rest'),outfile))
return True
elif self.getStrLC('Rest'):
self.printLog('#OUTFMT','REST output format "%s" not recognised.' % self.getStrLC('Rest'))
if self.i() < 0 or not rje.yesNo('Output all parsed outputs?'): return False
outfile = '%s.rest' % rbase
open(outfile,'w').write(self.restFullOutput())
self.printLog('#OUT','full: %s' % (outfile))
return True
for rkey in outputs:
if rkey in self.dict['Outfile']:
rje.backup(self,self.dict['Outfile'][rkey])
open(self.dict['Outfile'][rkey],'w').write(self.dict['Output'][rkey])
self.printLog('#OUT','%s: %s' % (rkey,self.dict['Outfile'][rkey]))
elif rkey not in ['intro']: self.warnLog('No outfile parsed/generated for %s output' % rkey)
def saveReadMe(self,filename='pydocs.txt',append=False): ### Prints docs for modules to file
'''
Prints docs for modules to file.
>> filename:str = output file name
>> append:boolean
'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
pydoc = self.obj['PyDoc']
if append:
self.printLog('#DOC','Appending docstrings to %s' % filename)
PYDOC = open(filename,'a')
else:
rje.mkDir(self,filename)
self.printLog('#DOC','Writing docstrings to %s' % filename)
PYDOC = open(filename,'w')
PYDOC.write(self.readMeHeader())
db = self.db('Module')
dx = 0
### ~ [2] Output Docstrings ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for sourcedir in pydoc.list['SourceDir']:
PYDOC.write('-%s:\n\n' % sourcedir)
for pyfile in db.dataKeys():
entry = db.data(pyfile)
module = entry['Module']
if not pyfile.find(sourcedir) >= 0 or not os.path.exists('%s%s%s.py' % (pydoc.getStr('PyPath'),rje.makePath(sourcedir),module)): continue
## ~ [2a] ~ Module docstring ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
mtxt = '### ~~~ Module %s ~ [%s] ~~~ ###' % (module,pyfile)
while len(mtxt) < 122: mtxt = mtxt[:5] + '~' + mtxt[5:-5] + '~' + mtxt[-5:]
try: PYDOC.write('%s\n\n%s\n' % (mtxt,entry['DocString'])); dx += 1
except:
self.errorLog('Cannot write DocString for %s' % module,printerror=False)
PYDOC.write('%s\n\nDocString Error!\n' % (mtxt)); dx += 1
PYDOC.write('\n\n\n')
PYDOC.close()
self.printLog('#DOC','Output to %s complete: %s modules.' % (filename,rje.iStr(dx)))
except: self.errorLog('Error in %s.saveDocs()' % self.prog())
def mapPhosByBLAST(self,fasfile): ### BLAST sequences against phosphoDB, align hits & mark sites (ID & Homology)
'''BLAST sequences against phosphoDB, align hits and mark phosphosites (ID & Homology).'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
## ~ [1a] Setup fasfile ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
scmd = self.cmd_list + ['seqin=%s' % fasfile,'autoload=T','autofilter=F']
qseqlist = rje_seq.SeqList(self.log,scmd)
qdict = qseqlist.seqNameDic()
## ~ [1b] Setup results files/directories ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
basefile = rje.baseFile(fasfile)
if self.info['PhosRes'].lower() in ['','none']: self.info['PhosRes'] = '%s.phosres.tdt' % basefile
headers = ['Name','Pos','AA','PELM','PELMPos','Evidence']
delimit = rje.getDelimit(self.cmd_list,rje.delimitFromExt(filename=self.info['PhosRes']))
rje.delimitedFileOutput(self,self.info['PhosRes'],headers,delimit,rje_backup=True)
ppath = rje.makePath('PhosALN')
rje.mkDir(self,ppath)
## ~ [1c] Setup BLAST ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
pblast = rje_blast.BLASTRun(self.log,self.cmd_list+['formatdb=F'])
pblast.setInfo({'Name':'%s.p.blast' % rje.baseFile(fasfile),'DBase':self.info['PELMFas'],'InFile':fasfile})
pblast.setStat({'HitAln':pblast.stat['OneLine']})
pblast.opt['Complexity Filter'] = False
pblast.formatDB(force=False)
## ~ [1d] Setup GABLAM Stats ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
gkey = 'GABLAMO ID' #x# % self.info['GABLAMO Key']
for g in ['ID','Hom']:
if self.stat['%sSim' % g] < 1.0: self.stat['%sSim' % g] *= 100.0
self.stat['%sSim' % g] = max(0.0,self.stat['%sSim' % g])
### ~ [2] PhosphoBLAST ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
pblast.blast(use_existing=True,log=True) # BLAST
pblast.readBLAST(gablam=True) # Read in
while pblast.search:
## ~ [2a] Align relevant hits from each BLAST ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
search = pblast.search.pop(0)
qseq = qdict[search.info['Name']]
idlist = []
qlen = qseq.aaLen()
hitdict = search.hitSeq(self.obj['SeqList'])
aln = rje_seq.SeqList(self.log,self.cmd_list+['autoload=F','autofilter=F'])
aln.seq = [qseq]
pdict = {} # Dictionary of {hseq:[poslist]}
rdict = {qseq:0} # Dictionary of {hseq:res}
for hit in search.hit[0:]:
hseq = hitdict[hit]
pdict[hseq] = []
for pos in rje.sortKeys(self.dict['PhosphoSites'][hseq.info['AccNum']]): pdict[hseq].append(pos)
if hit.info['Name'] == search.info['Name']:
if qseq.getSequence(case=False,gaps=False) != hseq.getSequence(case=False,gaps=False):
self.log.errorLog('Major problem: Search/Hit sequence mismatch for same sequence "%s"' % hit.info['Name'])
idlist.append(qseq)
pdict[qseq] = pdict.pop(hseq)
continue
gdict = hit.globalFromLocal(qlen)
qvh = float(100 * gdict['Query'][gkey]) / float(qlen)
if qvh < self.stat['HomSim']:
pdict.pop(hseq)
continue
aln.seq.append(hseq)
if (qseq.sameSpec(hseq) or not self.opt['UseSpec']) and qvh >= self.stat['IDSim']: idlist.append(hseq)
rdict[hseq] = 0
aln.muscleAln() #x#outfile='%s%s.phosaln.fas' % (ppath,qseq.info['AccNum']))
aln._addSeq('PhosAln','-' * qseq.seqLen())
aln.info['Name'] = '%s%s.phosaln.fas' % (ppath,qseq.info['AccNum'])
## ~ [2b] Map phosphorylations ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
print '>>>\n', aln.seq, pdict.keys(), rdict.keys()
for a in range(qseq.seqLen()):
if qseq.info['Sequence'][a] != '-': rdict[qseq] += 1
for hseq in pdict:
if hseq.info['Sequence'][a] == '-': continue
if hseq != qseq: rdict[hseq] += 1
if rdict[hseq] in pdict[hseq] and qseq.info['Sequence'][a] == hseq.info['Sequence'][a]: # Phosphosite
pdata = {'Name':search.info['Name'],'Pos':rdict[qseq],'AA':qseq.info['Sequence'][a],
'PELM':hseq.shortName(),'PELMPos':rdict[hseq],'Evidence':'Hom'}
if hseq == qseq: pdata['Evidence'] = 'Self'
elif hseq in idlist: pdata['Evidence'] = 'ID'
rje.delimitedFileOutput(self,self.info['PhosRes'],headers,delimit,pdata)
self.addPhos(aln.seq[-1],a,pdata['Evidence'])
## ~ [2c] Add Scansite/NetPhos if made? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
## ~ [2d] Save alignment ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
aln.saveFasta()
# Align hits for each > X %ID
# Map phosphosites onto alignment and output #
return
except: self.log.errorLog('Problem during PhosphoSeq.mapPhosByBLAST')
def mapPhosByBLAST(
self, fasfile
): ### BLAST sequences against phosphoDB, align hits & mark sites (ID & Homology)
'''BLAST sequences against phosphoDB, align hits and mark phosphosites (ID & Homology).'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
## ~ [1a] Setup fasfile ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
scmd = self.cmd_list + [
'seqin=%s' % fasfile, 'autoload=T', 'autofilter=F'
]
qseqlist = rje_seq.SeqList(self.log, scmd)
qdict = qseqlist.seqNameDic()
## ~ [1b] Setup results files/directories ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
basefile = rje.baseFile(fasfile)
if self.info['PhosRes'].lower() in ['', 'none']:
self.info['PhosRes'] = '%s.phosres.tdt' % basefile
headers = ['Name', 'Pos', 'AA', 'PELM', 'PELMPos', 'Evidence']
delimit = rje.getDelimit(
self.cmd_list,
rje.delimitFromExt(filename=self.info['PhosRes']))
rje.delimitedFileOutput(self,
self.info['PhosRes'],
headers,
delimit,
rje_backup=True)
ppath = rje.makePath('PhosALN')
rje.mkDir(self, ppath)
## ~ [1c] Setup BLAST ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
pblast = rje_blast.BLASTRun(self.log,
self.cmd_list + ['formatdb=F'])
pblast.setInfo({
'Name': '%s.p.blast' % rje.baseFile(fasfile),
'DBase': self.info['PELMFas'],
'InFile': fasfile
})
pblast.setStat({'HitAln': pblast.stat['OneLine']})
pblast.opt['Complexity Filter'] = False
pblast.formatDB(force=False)
## ~ [1d] Setup GABLAM Stats ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
gkey = 'GABLAMO ID' #x# % self.info['GABLAMO Key']
for g in ['ID', 'Hom']:
if self.stat['%sSim' % g] < 1.0:
self.stat['%sSim' % g] *= 100.0
self.stat['%sSim' % g] = max(0.0, self.stat['%sSim' % g])
### ~ [2] PhosphoBLAST ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
pblast.blast(use_existing=True, log=True) # BLAST
pblast.readBLAST(gablam=True) # Read in
while pblast.search:
## ~ [2a] Align relevant hits from each BLAST ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
search = pblast.search.pop(0)
qseq = qdict[search.info['Name']]
idlist = []
qlen = qseq.aaLen()
hitdict = search.hitSeq(self.obj['SeqList'])
aln = rje_seq.SeqList(
self.log, self.cmd_list + ['autoload=F', 'autofilter=F'])
aln.seq = [qseq]
pdict = {} # Dictionary of {hseq:[poslist]}
rdict = {qseq: 0} # Dictionary of {hseq:res}
for hit in search.hit[0:]:
hseq = hitdict[hit]
pdict[hseq] = []
for pos in rje.sortKeys(
self.dict['PhosphoSites'][hseq.info['AccNum']]):
pdict[hseq].append(pos)
if hit.info['Name'] == search.info['Name']:
if qseq.getSequence(case=False,
gaps=False) != hseq.getSequence(
case=False, gaps=False):
self.log.errorLog(
'Major problem: Search/Hit sequence mismatch for same sequence "%s"'
% hit.info['Name'])
idlist.append(qseq)
pdict[qseq] = pdict.pop(hseq)
continue
gdict = hit.globalFromLocal(qlen)
qvh = float(100 * gdict['Query'][gkey]) / float(qlen)
if qvh < self.stat['HomSim']:
pdict.pop(hseq)
continue
aln.seq.append(hseq)
if (qseq.sameSpec(hseq) or not self.opt['UseSpec']
) and qvh >= self.stat['IDSim']:
idlist.append(hseq)
rdict[hseq] = 0
def gopher(self): ### Sets up data for GOPHER run
'''Sets up data for GOPHER run.'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
rje.mkDir(self, 'BLAST/')
rje_blast.BLASTRun(self.log, self.cmd_list).formatDB(
fasfile='%s.ygob.fas' % self.info['Basefile'],
protein=True,
force=False)
rje_blast.BLASTRun(self.log, self.cmd_list).formatDB(
fasfile='%s.yeast.fas' % self.info['Basefile'],
protein=True,
force=False)
seqdict = self.obj['SeqList'].seqNameDic('AccNum')
ymap = self.dict['PillarMap'] = {}
### ~ [2] Convert Pillars to BLAST IDs ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
(px, ptot) = (0.0, len(self.list['Pillars']))
ox = 0
for pillar in self.list['Pillars']:
self.progLog(
'\r#YGOB',
'Converting YGOB Pillars for GOPHER: %.2f%%' % (px / ptot))
px += 100
newpillar = []
for yid in pillar:
seq = rje_sequence.Sequence(self.log, self.cmd_list)
seq.opt['Yeast'] = True
#self.deBug(yid)
seq.info['Name'] = yid
seq.extractDetails(gnspacc=True)
#self.deBug(seq.info)
ygob = seq.info['AccNum']
if ygob in self.dict['Rename']:
acc = self.dict['Rename'][ygob]
else:
acc = ygob
ymap[yid] = acc
if acc not in seqdict:
self.printLog(
'\r#GENE',
'Non-coding gene %s (%s)? Cannot find in fasta file'
% (acc, yid))
continue
try:
newpillar.append(seqdict[acc].shortName())
except:
print yid, ygob, acc
self.errorLog(rje_zen.Zen().wisdom())
if not newpillar: continue
for ygob in pillar:
acc = ymap[ygob]
if acc not in seqdict: continue
if acc in self.list['YeastSeq'] or (
not self.list['YeastSeq']
and seqdict[acc].info['SpecCode'] == 'YEAST'):
open(
rje.makePath('BLAST/%s.blast.id' % acc,
wholepath=True),
'w').write(string.join(newpillar, '\n'))
ox += 1
self.progLog(
'\r#YGOB',
'Converted YGOB Pillars for GOPHER: %s BLAST ID files.' %
rje.iStr(ox))
except:
self.errorLog(rje_zen.Zen().wisdom())
raise # Delete this if method error not terrible
def fpi(self): ### Family-protein interactions
'''Family-protein interactions.'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
if not self.dict['Domain']: return
outdir = 'SLiMPID_FPI'
rje.mkDir(self, outdir)
fpi = {} # Dictionary of {family:[interactors]}
badname = []
### ~ [2] Process ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
for qry in rje.sortKeys(self.dict['PPI']):
try:
fam = self.dict['Fam'][qry]
if len(fam) < 2: continue
except:
self.errorLog('Problem with "%s" protein family' % qry)
continue
fpi[qry] = []
for hub in fam:
if hub not in self.dict['PPI']: continue
fpi[qry] += self.dict['PPI'][hub] # Add with redundancy
for spoke in fpi[qry][0:]:
if fpi[qry].count(spoke) == 1:
fpi[qry].remove(
spoke) # Must have 2+ family interactions
for hub in fam:
if hub not in self.dict['PPI']: continue
for spoke in self.dict['PPI'][hub][0:]:
if spoke in fpi[qry]:
self.dict['PPI'][hub].remove(spoke)
if spoke in self.dict['PPI'] and hub in self.dict[
'PPI'][spoke]:
self.dict['PPI'][spoke].remove(hub)
fpi[qry] = rje.sortUnique(fpi[qry], False, False)
acc = []
gene = self.dict['Gene'][qry]
for name in fpi[qry]:
if not name: continue
if name in self.dict['Seq']:
acc.append(self.dict['Seq'][name].info['AccNum'])
elif name not in badname:
badname.append(name)
open('%s/%s.fpi.acc' % (outdir, gene),
'w').write(string.join(acc, '\n'))
self.printLog('#FPI',
'%s family => %d interactors' % (gene, len(acc)))
if badname:
badname.sort()
self.printLog(
'#BAD', '%d "bad" protein names: %s' %
(len(badname), string.join(badname, '; ')))
### ~ [3] Cleanup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
hx = len(self.dict['PPI'])
for hub in rje.sortKeys(self.dict['PPI']):
if hub and self.dict['PPI'][hub]: continue
self.dict['PPI'].pop(hub)
self.printLog('#FPI', 'No %s PPI left after FPI removed' % hub)
self.printLog(
'#PPX', '%s of %s PPI hubs remain after FPI removed' %
(rje.integerString(len(
self.dict['PPI'])), rje.integerString(hx)))
except:
self.errorLog('Problem with SLiMPID.fpi()', quitchoice=True)
def mapRegionsToSequences(
self): ### Maps tabulates PPI regions onto sequence datasets
'''Maps tabulates PPI regions onto sequence datasets.'''
try: ### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
minseq = 3
outdir = 'RegPPI/'
adddir = 'RegPPIAdd/'
rje.mkDir(self, outdir)
rje.mkDir(self, adddir)
tabfile = 'ppi_region.tdt'
region = rje.dataDict(self,
tabfile, ['Interactor', 'Protein'],
['Start', 'End'],
lists=True)
### ~ [2] Work through each pair in turn ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
px = 0.0
ptot = len(region)
fx = 0
for pair in rje.sortKeys(region):
self.progLog('\r#FAS',
'Generating fasta files: %.2f%%' % (px / ptot))
px += 100.0
## ~ [2a] Map sequences to PPI dictionary ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
[hub, spoke] = string.split(pair, '\t')
try:
qryseq = self.dict['SeqObj'][spoke]
except:
self.printLog(
'\n#QRY',
'Spoke gene "%s" missing from Sequence file' % spoke)
continue
try:
spoke = self.dict['Seq2Gene'][spoke]
except:
self.printLog(
'\n#QRY',
'Spoke protein "%s" missing from PPI dictionary' %
spoke)
continue
if hub not in self.dict['PPI']:
self.printLog(
'\n#HUB',
'Hub gene "%s" missing from PPI dictionary' % hub)
continue
addspoke = spoke not in self.dict['PPI'][hub]
if addspoke:
self.dict['PPI'][hub].append(spoke)
self.printLog(
'\n#PPI',
'Added spoke gene "%s" to hub "%s" interactome' %
(spoke, hub))
if len(self.dict['PPI'][hub]) < minseq:
self.printLog(
'\n#HUB',
'Hub "%s" interactome too small (<%s spokes)' %
(hub, minseq))
continue
## ~ [2b] Identify query sequence ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
reglist = []
for pos in region[pair]['Start'] + region[pair]['End']:
reglist.append(string.atoi(pos))
reglist.sort()
qsequence = qryseq.info['Sequence'].lower()
self.deBug(len(qsequence))
self.deBug(qsequence)
prelen = len(qsequence)
while reglist:
self.deBug(reglist)
try:
startx = reglist.pop(0) - 1
endx = reglist.pop(0)
except:
self.errorLog('%s PPI Region problem: %s' %
(pair, region[pair]))
continue
self.deBug(qsequence[startx - 1:endx + 1].upper())
qsequence = qsequence[:startx] + qsequence[
startx:endx].upper() + qsequence[endx:]
self.deBug(qsequence)
if len(qsequence) != prelen:
self.printLog('#F**K', '%s' % region[pair])
self.printLog('#F**K', qryseq.info['Sequence'].lower())
self.printLog('#F**K', qsequence)
raise ValueError
## ~ [2c] Output sequences ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
if addspoke:
outfile = '%s%s.%s.fas' % (adddir, hub, spoke)
ox = 1
else:
outfile = '%s%s.%s.fas' % (outdir, hub, spoke)
ox = 1
open(outfile,
'w').write('>%s\n%s\n' % (qryseq.info['Name'], qsequence))
for spoke2 in self.dict['PPI'][hub]:
if spoke2 == spoke: continue
try:
sseq = self.dict['SeqObj'][self.dict['Gene2Seq']
[spoke2]]
open(outfile, 'a').write(
'>%s\n%s\n' %
(sseq.info['Name'], sseq.info['Sequence']))
ox += 1
except:
pass
self.printLog('\n#FAS',
'%s sequences output to %s' % (ox, outfile))
except:
self.errorLog(rje_zen.Zen().wisdom())
raise # Delete this if method error not terrible
def domainFasta(
self
): ### Outputs parsed domain and domain PPI datasets in Fasta format
'''Outputs parsed PPI datasets in Fasta format.'''
try:
### ~ Tab delimited domain-HPRD pairs ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
headers = ['Domain', 'HPRD', 'Gene']
dfile = self.info['OutDir'] + 'HPRD.domains.tdt'
rje.delimitedFileOutput(self, dfile, headers, '\t')
sfile = self.info['OutDir'] + 'HPRD.domsource.tdt'
shead = ['Domain', 'Source']
rje.delimitedFileOutput(self, sfile, shead, '\t')
dx = 0.0
for domain in rje.sortKeys(self.dict['Domains']):
self.log.printLog('\r#DOM',
'HPRD Domain output (%s): %.1f%%' %
(dfile, dx / len(self.dict['Domains'])),
newline=False,
log=False)
dx += 100.0
for hid in self.dict['Domains'][domain]:
datadict = {
'Domain': domain,
'HPRD': hid,
'Gene': self.dict['HPRD'][hid]['gene']
}
rje.delimitedFileOutput(self, dfile, headers, '\t',
datadict)
for source in self.dict['DomainSource'][domain]:
datadict = {'Domain': domain, 'Source': source}
rje.delimitedFileOutput(self, sfile, shead, '\t', datadict)
self.log.printLog(
'\r#DOM', 'HPRD Domain output (%s): %s domains.' %
(dfile, rje.integerString(len(self.dict['Domains']))))
### ~ Domain PPI Dataset Outputs ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
datpath = self.info['OutDir'] + rje.makePath(
'HPRD_Domain_Datasets/')
rje.mkDir(self, datpath)
for domain in rje.sortKeys(self.dict['Domains']):
## Generate a list of all interactors with domain-containing proteins ##
plist = []
for p1 in self.dict['Domains'][domain]:
if p1 not in self.dict['PPI']: continue
for p2 in self.dict['PPI'][p1]:
if p2 not in plist: plist.append(p2)
plist.sort()
## Generate Sequence list and output ##
mylist = []
for p in plist:
if self.opt['AllIso']:
mylist += self.dict['HPRD'][p]['Seq']
else:
mylist.append(self.dict['HPRD'][p]['Seq'])
sfile = '%s%s_hprd.fas' % (datpath, domain)
if mylist:
self.obj['SeqList'].saveFasta(seqs=mylist, seqfile=sfile)
else:
self.log.printLog(
'#DOM', 'No PPI partners for domain "%s"' % domain)
self.log.printLog('\r#DOM', 'HPRD Domain fasta output complete.')
except:
self.log.errorLog('Error in HPRD.saveFasta()',
printerror=True,
quitchoice=False)
raise
def mapRegionsToSequences(self): ### Maps tabulates PPI regions onto sequence datasets
'''Maps tabulates PPI regions onto sequence datasets.'''
try:### ~ [1] Setup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
minseq = 3
outdir = 'RegPPI/'
adddir = 'RegPPIAdd/'
rje.mkDir(self,outdir)
rje.mkDir(self,adddir)
tabfile = 'ppi_region.tdt'
region = rje.dataDict(self,tabfile,['Interactor','Protein'],['Start','End'],lists=True)
### ~ [2] Work through each pair in turn ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ###
px = 0.0; ptot = len(region); fx = 0
for pair in rje.sortKeys(region):
self.progLog('\r#FAS','Generating fasta files: %.2f%%' % (px/ptot)); px += 100.0
## ~ [2a] Map sequences to PPI dictionary ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
[hub, spoke] = string.split(pair,'\t')
try: qryseq = self.dict['SeqObj'][spoke]
except: self.printLog('\n#QRY','Spoke gene "%s" missing from Sequence file' % spoke); continue
try: spoke = self.dict['Seq2Gene'][spoke]
except: self.printLog('\n#QRY','Spoke protein "%s" missing from PPI dictionary' % spoke); continue
if hub not in self.dict['PPI']: self.printLog('\n#HUB','Hub gene "%s" missing from PPI dictionary' % hub); continue
addspoke = spoke not in self.dict['PPI'][hub]
if addspoke:
self.dict['PPI'][hub].append(spoke)
self.printLog('\n#PPI','Added spoke gene "%s" to hub "%s" interactome' % (spoke,hub))
if len(self.dict['PPI'][hub]) < minseq: self.printLog('\n#HUB','Hub "%s" interactome too small (<%s spokes)' % (hub,minseq)); continue
## ~ [2b] Identify query sequence ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
reglist = []
for pos in region[pair]['Start'] + region[pair]['End']: reglist.append(string.atoi(pos))
reglist.sort()
qsequence = qryseq.info['Sequence'].lower()
self.deBug(len(qsequence))
self.deBug(qsequence)
prelen = len(qsequence)
while reglist:
self.deBug(reglist)
try: startx = reglist.pop(0) - 1; endx = reglist.pop(0)
except: self.errorLog('%s PPI Region problem: %s' % (pair,region[pair])); continue
self.deBug(qsequence[startx-1:endx+1].upper())
qsequence = qsequence[:startx] + qsequence[startx:endx].upper() + qsequence[endx:]
self.deBug(qsequence)
if len(qsequence) != prelen:
self.printLog('#F**K','%s' % region[pair])
self.printLog('#F**K',qryseq.info['Sequence'].lower())
self.printLog('#F**K',qsequence)
raise ValueError
## ~ [2c] Output sequences ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ##
if addspoke: outfile = '%s%s.%s.fas' % (adddir,hub,spoke); ox = 1
else: outfile = '%s%s.%s.fas' % (outdir,hub,spoke); ox = 1
open(outfile,'w').write('>%s\n%s\n' % (qryseq.info['Name'],qsequence))
for spoke2 in self.dict['PPI'][hub]:
if spoke2 == spoke: continue
try:
sseq = self.dict['SeqObj'][self.dict['Gene2Seq'][spoke2]]
open(outfile,'a').write('>%s\n%s\n' % (sseq.info['Name'],sseq.info['Sequence']))
ox += 1
except: pass
self.printLog('\n#FAS','%s sequences output to %s' % (ox,outfile))
except:
self.errorLog(rje_zen.Zen().wisdom())
raise # Delete this if method error not terrible
