def dbCleanByPattern( pattern, inFileName, outFileName="", verbose=0 ):
if pattern == "":
return
patternToSearch = re.compile(pattern)
if outFileName == "":
outFileName = inFileName + '.cleaned'
outFile = open(outFileName,'w')
bioseq = Bioseq()
bioseqNb = 0
savedBioseqNb = 0
inFile = open(inFileName)
while True:
bioseq.read(inFile)
if bioseq.sequence == None:
break
bioseqNb += 1
if not patternToSearch.search(bioseq.header):
bioseq.write(outFile)
if verbose > 1:
print 'sequence num',bioseqNb,'[',bioseq.header[0:40],'...] saved !!'
savedBioseqNb += 1
inFile.close()
outFile.close()
if verbose > 0:
print "%i sequences saved in file '%s'" % ( savedBioseqNb, outFileName )
def spliceFromCoords( genomeFile, coordFile, obsFile ):
genomeFileHandler = open( genomeFile, "r" )
obsFileHandler = open( obsFile, "w" )
dChr2Maps = MapUtils.getDictPerSeqNameFromMapFile( coordFile )
while True:
bs = Bioseq()
bs.read( genomeFileHandler )
if bs.sequence == None:
break
if dChr2Maps.has_key( bs.header ):
lCoords = MapUtils.getMapListSortedByIncreasingMinThenMax( dChr2Maps[ bs.header ] )
splicedSeq = ""
currentSite = 0
for iMap in lCoords:
minSplice = iMap.getMin() - 1
if minSplice > currentSite:
splicedSeq += bs.sequence[ currentSite : minSplice ]
currentSite = iMap.getMax()
splicedSeq += bs.sequence[ currentSite : ]
bs.sequence = splicedSeq
bs.write( obsFileHandler )
genomeFileHandler.close()
obsFileHandler.close()
def dbExtractByPattern( pattern, inFileName, outFileName="", verbose=0 ):
if pattern == "":
return
if outFileName == "":
outFileName = inFileName + '.extracted'
outFile = open( outFileName, 'w' )
patternTosearch = re.compile( pattern )
bioseq = Bioseq()
bioseqNb = 0
savedBioseqNb = 0
inFile = open( inFileName, "r" )
while True:
bioseq.read( inFile )
if bioseq.sequence == None:
break
bioseqNb = bioseqNb + 1
m = patternTosearch.search( bioseq.header )
if m:
bioseq.write( outFile )
if verbose > 1:
print 'sequence num',bioseqNb,'matched on',m.group(),'[',bioseq.header[0:40],'...] saved !!'
savedBioseqNb = savedBioseqNb + 1
inFile.close()
outFile.close()
if verbose > 0:
print "%i sequences saved in file '%s'" % ( savedBioseqNb, outFileName )
def dbLengthFilter( len_min, inFileName, verbose=0 ):
file_db = open( inFileName, "r" )
file_dbInf = open( inFileName+".Inf"+str(len_min), "w" )
file_dbSup = open( inFileName+".Sup"+str(len_min), "w" )
seq = Bioseq()
numseq = 0
nbsave = 0
while True:
seq.read( file_db )
if seq.sequence == None:
break
l = seq.getLength()
numseq = numseq + 1
if l >= len_min:
seq.write( file_dbSup )
if verbose > 0:
print 'sequence #',numseq,'=',l,'[',seq.header[0:40],'...] Sup !!'
nbsave=nbsave+1
else:
seq.write( file_dbInf )
if verbose > 0:
print 'sequence #',numseq,'=',l,'[',seq.header[0:40],'...] Inf !!'
nbsave=nbsave+1
file_db.close()
file_dbInf.close()
file_dbSup.close()
if verbose > 0:
print nbsave,'saved sequences in ',inFileName+".Inf"+str(len_min)," and ", inFileName+".Sup"+str(len_min)
def dbCleanByFilePattern( patternFileName, inFileName, outFileName="", verbose=0 ):
if patternFileName == "":
print "ERROR: no file of pattern"
sys.exit(1)
bioseq = Bioseq()
bioseqNb = 0
savedBioseqNb = 0
lHeaders = []
inFile = open( inFileName, "r" )
while True:
bioseq.read( inFile )
if bioseq.sequence == None:
break
bioseqNb += 1
lHeaders.append( bioseq.header )
inFile.close()
patternFile = open( patternFileName, "r")
lHeadersToRemove = []
for pattern in patternFile:
if verbose > 0:
print "pattern: ",pattern[:-1]; sys.stdout.flush()
patternToSearch = re.compile( pattern[:-1] )
for h in lHeaders:
if patternToSearch.search(h):
lHeadersToRemove.append(h)
patternFile.close()
if outFileName == "":
outFileName = inFileName + '.cleaned'
outFile = open( outFileName, 'w' )
bioseqNum = 0
inFile=open( inFileName )
while True:
bioseq.read( inFile )
bioseqNum += 1
if bioseq.sequence == None:
break
if bioseq.header not in lHeadersToRemove:
bioseq.write( outFile )
if verbose > 1:
print 'sequence num',bioseqNum,'/',bioseqNb,'[',bioseq.header[0:40],'...] saved !!'; sys.stdout.flush()
savedBioseqNb += 1
inFile.close()
outFile.close()
if verbose > 0:
print "%i sequences saved in file '%s'" % ( savedBioseqNb, outFileName )
def dbExtractByFilePattern( patternFileName, inFileName, outFileName="", verbose=0 ):
if patternFileName == "":
print "ERROR: no file of pattern"
sys.exit(1)
bioseq = Bioseq()
bioseqNb = 0
savedBioseqNb = 0
lHeaders = []
inFile = open( inFileName, "r" )
while True:
bioseq.read( inFile )
if bioseq.sequence == None:
break
lHeaders.append( bioseq.header )
inFile.close()
lHeadersToKeep = []
patternFile = open( patternFileName, "r" )
for pattern in patternFile:
if verbose > 0:
print "pattern: ",pattern[:-1]; sys.stdout.flush()
patternToSearch = re.compile(pattern[:-1])
for h in lHeaders:
if patternToSearch.search(h):
lHeadersToKeep.append(h)
patternFile.close()
if outFileName == "":
outFileName = inFileName + ".extracted"
outFile=open( outFileName, "w" )
inFile = open( inFileName, "r" )
while True:
bioseq.read(inFile)
if bioseq.sequence == None:
break
bioseqNb += 1
if bioseq.header in lHeadersToKeep:
bioseq.write(outFile)
if verbose > 1:
print 'sequence num',bioseqNb,'[',bioseq.header[0:40],'...] saved !!'; sys.stdout.flush()
savedBioseqNb += 1
inFile.close()
outFile.close()
if verbose > 0:
print "%i sequences saved in file '%s'" % ( savedBioseqNb, outFileName )
def filterClassifiedConsensus( self ):
inFile = open( self._inFaFile, "r" )
outFile = open( self._outFaFile, "w" )
bs = Bioseq()
nbInSeq = 0
nbRmv = 0
if self._classifFile != "":
dHeader2Classif = self.getClassifPerHeaderOfUnclassifiedConsensus()
while True:
bs.read( inFile )
if bs.header == None:
break
nbInSeq += 1
if self._verbose > 1:
print bs.header
if self._filterSSRs and "SSR" in bs.header and ( self._maxLengthToFilterSSRs == 0 or bs.getLength() <= self._maxLengthToFilterSSRs ):
nbRmv += 1
if self._verbose > 1: print "filtered SSR !"
elif self._filterHostGenes and "HostGene" in bs.header:
nbRmv += 1
if self._verbose > 1: print "filtered HostGene !"
elif self._filterConfused and "confused" in bs.header and "confusedness=no" not in bs.header:
nbRmv += 1
if self._verbose > 1: print "filtered confused !"
elif self._filterNoCat != "0" and "NoCat" in bs.header:
keep = False
if "2" in self._filterNoCat:
algoMSA = ""
for i in ["Map","MAP","Malign","Mafft","Prank","Clustalw","Muscle","Tcoffee"]:
if i in bs.header:
algoMSA = i
nbAlignSeq = int( bs.header.split(algoMSA+"_")[1].split("|")[0] )
if nbAlignSeq > self._nbAlignSeqNoCat:
keep = True
if "3" in self._filterNoCat:
for header in dHeader2Classif.keys():
if header in bs.header:
if "no structural features" not in dHeader2Classif[header][6]:
keep = True
if keep:
bs.write( outFile )
else:
nbRmv += 1
if self._verbose > 1: print "filtered NoCat !"
elif self._filterIncomplete and "completeness=incomp" in bs.header:
nbRmv += 1
if self._verbose > 1: print "filtered incomplete !"
else:
bs.write( outFile )
inFile.close()
outFile.close()
if self._verbose > 0:
print "nb of input seq: %i" % ( nbInSeq )
print "nb of filtered seq: %i" % ( nbRmv )
sys.stdout.flush()
