def dbSumRep(inFileName):
"""
deprecated
"""
import reputer
stat={"F":Stat(),"R":Stat(),"C":Stat(),"P":Stat()}
file_db=open(inFileName)
seq=Bioseq()
numseq=0
while 1:
seq.read(file_db)
if seq.sequence==None:
break
hist=reputer.dist(seq,"-allmax -f -r -c -p -l 3")
sumL={"F":0,"R":0,"C":0,"P":0}
sumN={"F":0,"R":0,"C":0,"P":0}
for i in hist:
sumL[i[0]]=sumL[i[0]]+(i[1]*i[2])
sumN[i[0]]=sumN[i[0]]+i[2]
for i in sumL.keys():
sumL[i]=-(math.log(float(sumL[i])/sumN[i])-math.log(seq.getLength()))
stat[i].add(sumL[i])
numseq=numseq+1
print 'sequence #',numseq,'=',seq.getLength(),'[',\
seq.header[0:40],'...]'
print "\tF=%5.3f R=%5.3f C=%5.3f P=%5.3f" % \
(sumL["F"],sumL["R"],sumL["C"],sumL["P"])
file_db.close()
for i in stat.keys():
print i,"=>",stat[i].string()
def dbORF( inFileName, nb=0, size=0, outFileName="" ):
inFile = open( inFileName )
if outFileName == "":
outFileName = inFileName + ".orf.map"
outFile = open( outFileName, "w" )
seq = Bioseq()
seq_out = Bioseq()
numseq = 0
while 1:
seq.read( inFile )
if seq.sequence == None:
break
seq.upCase()
numseq = numseq + 1
print 'sequence #',numseq,'=',seq.getLength(),'[',seq.header[0:40],'...]'
orf = seq.findORF()
best_orf = []
for i in orf.keys():
l = len(orf[i])
for j in xrange(1,l):
start = orf[i][j-1] + 4
end = orf[i][j] + 3
if end - start >= size:
best_orf.append( ( end-start, i+1, start, end ) )
seq.sequence = seq.complement()
orf = seq.findORF()
seqlen = seq.getLength()
for i in orf.keys():
l = len(orf[i])
for j in xrange(1,l):
start = seqlen - orf[i][j-1] - 3
end = seqlen - orf[i][j] - 2
if start - end >= size:
best_orf.append( ( start-end, (i+1)*-1, start, end ) )
best_orf.sort()
best_orf.reverse()
l = len(best_orf)
if nb > l or nb == 0 :
nb = l
for i in xrange(0,nb):
print best_orf[i]
outFile.write("%s\t%s\t%d\t%d\n"%("ORF|"+str(best_orf[i][1])+\
"|"+str(best_orf[i][0]),seq.header,
best_orf[i][2],best_orf[i][3]))
inFile.close()
outFile.close()
return 0
def dbRelEntropy(inFileName,wordsize):
"""
deprecated
"""
file_db=open(inFileName)
seq=Bioseq()
refocc={}
sumlen=0
while 1:
seq.read(file_db)
if seq.sequence==None:
break
sumlen=sumlen+seq.getLength()-wordsize
occ=seq.occ_word(wordsize)
if(len(refocc)==0):
refocc=occ
else:
for w in occ.keys():
if refocc.has_key(w):
refocc[w]=refocc[w]+occ[w]
else:
refocc[w]=occ[w]
file_db.close()
reffreq={}
for w in refocc.keys():
reffreq[w]=float(refocc[w]+1)/sumlen
vec_len=[]
stat=Stat()
file_db=open(inFileName)
numseq=0
while 1:
seq.read(file_db)
if seq.sequence==None:
break
i=seq.rel_entropy(reffreq)
stat.add(i)
numseq=numseq+1
print 'sequence #',numseq,'=',seq.getLength(),'[',seq.header[0:40],'...] entropy',i
vec_len.append((i,numseq,seq.header))
file_db.close()
vec_len.sort()
for s in vec_len:
print 'H=',s[0],'=> #',s[1], s[2]
print stat.string()
return vec_len
def dbLengthFilter(len_min,inFileName, verbose=0): file_db=open(inFileName) 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 1: 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 dbEntropy(inFileName,wordsize): """ deprecated """ vec_len=[] stat=Stat() file_db=open(inFileName) seq=Bioseq() numseq=0 while 1: seq.read(file_db) if seq.sequence==None: break i=seq.entropy(wordsize) stat.add(i) numseq=numseq+1 print 'sequence #',numseq,'=',seq.getLength(),'[',seq.header[0:40],'...] entropy',i vec_len.append((-i,numseq,seq.header)) file_db.close() vec_len.sort() for s in vec_len: print 'I=',-s[0],'=> #',s[1], s[2] print stat.string() return vec_len
def dbWord(inFileName,wordsize):
"""
deprecated
"""
vec_len=[]
file_db=open(inFileName)
seq=Bioseq()
numseq=0
nb_word=0
statCount={}
while 1:
seq.read(file_db)
if seq.sequence==None:
break
numseq=numseq+1
print 'sequence #',numseq,'=',seq.getLength(),'[',seq.header[0:40],'...]'
occ,nb=seq.occ_word(wordsize)
nb_word=nb_word+nb
for i in occ.keys():
if i not in statCount.keys():
statCount[i]=occ[i]
else:
statCount[i]=statCount[i]+occ[i]
file_db.close()
vec_sort=[]
for i in statCount.keys():
vec_sort.append((-float(statCount[i])/nb_word,i))
vec_sort.sort()
for i in vec_sort:
print i[1],"=",-i[0]
def sortSequencesByIncreasingLength( inFile, outFile, verbose=0 ):
"""
Save sequences in 'inFile' into 'outFile' sorted by their length in increasing order.
"""
if verbose > 0:
print "sort sequences by increasing length"
sys.stdout.flush()
if not os.path.exists( inFile ):
print "ERROR: file '%s' doesn't exist" % ( inFile )
sys.exit(1)
# read each seq one by one
# save them in distinct temporary files
# with their length in the name
inFileHandler = open( inFile, "r" )
bs = Bioseq()
countSeq = 0
while True:
bs.read( inFileHandler )
if bs.header == None:
break
countSeq += 1
tmpFile = "%ibp_%inb" % ( bs.getLength(), countSeq )
bs.save( tmpFile )
if verbose > 1:
print "%s (%i bp) saved in '%s'" % ( bs.header, bs.getLength(), tmpFile )
bs.header = ""
bs.sequence = ""
inFileHandler.close()
# sort temporary file names
# concatenate them into the output file
if os.path.exists( outFile ):
os.remove( outFile )
lFiles = glob.glob( "*bp_*nb" )
lFiles.sort( key=lambda s:int(s.split("bp_")[0]) )
for fileName in lFiles:
cmd = "cat %s >> %s" % ( fileName, outFile )
returnValue = os.system( cmd )
if returnValue != 0:
print "ERROR while concatenating '%s' with '%s'" % ( fileName, outFile )
sys.exit(1)
os.remove( fileName )
return 0
def dbTraduit(inFileName,phase=0,complement='T',pep_filename=""): """ deprecated """ file_db=open(inFileName) if pep_filename=="": pep_filename=inFileName+'.pep' file_pep=open(pep_filename,'w') seq=Bioseq() seq_out=Bioseq() numseq=0 while 1: seq.read(file_db) if seq.sequence==None: break numseq=numseq+1 print 'sequence #',numseq,'=',seq.getLength(),\ '[',seq.header[0:40],'...]' if phase>=0 : if phase==1 or phase==0 : seq_out.sequence=seq.traduit(1) seq_out.header=seq.header+" (phase 1)" seq_out.write(file_pep) if phase==2 or phase==0 : seq_out.sequence=seq.traduit(2) seq_out.header=seq.header+" (phase 2)" seq_out.write(file_pep) if phase==3 or phase==0 : seq_out.sequence=seq.traduit(3) seq_out.header=seq.header+" (phase 3)" seq_out.write(file_pep) if complement=='T' or phase<0 : seq.sequence=seq.complement() if phase==-1 or phase==0 : seq_out.sequence=seq.traduit(1) seq_out.header=seq.header+" (phase -1)" seq_out.write(file_pep) if phase==-2 or phase==0 : seq_out.sequence=seq.traduit(2) seq_out.header=seq.header+" (phase -2)" seq_out.write(file_pep) if phase==-3 or phase==0 : seq_out.sequence=seq.traduit(3) seq_out.header=seq.header+" (phase -3)" seq_out.write(file_pep) file_db.close() file_pep.close()
def dbITRsearch(inFileName,len_min,mismatch,skip_len=20000): """ deprecated """ import reputer n=0 s=0 file_db=open(inFileName) file_out=open(inFileName+".stree_itr",'w') seq=Bioseq() numseq=0 while 1: seq.read(file_db) if seq.sequence==None: break numseq=numseq+1 print 'sequence #',numseq,'=',\ seq.getLength(),'[',\ seq.header[0:40],'...]' if seq.getLength()<skip_len: rep=reputer.find(seq,"-p -l "+str(len_min)+\ " -e "+str(mismatch)) for i in rep.rep_list: if i.pos1 < 5 \ and i.pos2+i.length2>seq.getLength()-5: i.view() n=n+1 seq.write(file_out) break else: s=s+1 print ' too long, skipped' print n,"found ", s, "skipped" file_db.close() file_out.close()
def db2map( inFileName, map_filename="" ): file_db = open( inFileName , "r" ) if map_filename == "": map_filename = inFileName + ".map" file_map = open( map_filename, "w" ) seq = Bioseq() numseq = 0 while 1: seq.read( file_db ) if seq.sequence == None: break numseq = numseq + 1 line='sequence'+str(numseq)+'\t'+seq.header+'\t1'+'\t'+str(seq.getLength()) print line file_map.write( line + "\n" ) file_db.close() file_map.close() print "saved in ",map_filename
def dbComplement(inFileName,comp_filename=""): """ deprecated """ file_db=open(inFileName) if comp_filename=="": comp_filename=inFileName+'.comp' file_comp=open(comp_filename,'w') seq=Bioseq() numseq=0 while 1: seq.read(file_db) if seq.sequence==None: break numseq=numseq+1 print 'sequence #',numseq,'=',seq.getLength(),'[',seq.header[0:40],'...]' seq.sequence=seq.complement() seq.header=seq.header+" (complement!)" seq.write(file_comp) file_db.close() file_comp.close()
def dbLength( inFileName ): """ return the length of each sequence in the input fasta file @param inFileName: name of the input fasta file @type inFileName: string """ if not os.path.exists( inFileName ): print "*** Error: %s doesn't exist" % ( inFileName ) sys.exit(1) vec_len = [] stat = Quantile() inFile = open( inFileName, "r" ) seq = Bioseq() numseq = 0 while 1: seq.read( inFile ) if seq.sequence == None: break l = seq.getLength() stat.add( l ) numseq = numseq + 1 print "sequence #",numseq,"=",l,"[",seq.header[0:40],"...]" vec_len.append( ( l, numseq, seq.header[0:40] ) ) inFile.close() vec_len.sort() for s in vec_len: print "len=",s[0],"=> #",s[1], s[2] print stat.string() print "total length=",stat.sum return vec_len
def dbConsensus(filename,consensus_filename,max_set_size=20,max_len=20000,min_len=50,min_base_nb=1):
"""
deprecated
"""
os.system("orienter "+filename)
tmp_consensus_filename=filename+".oriented.consensus.tmp"
size_db=dbSize(filename+".oriented")
file_in=open(filename+".oriented")
file_out=open(consensus_filename,'w')
seq=Bioseq()
if size_db==1:
seq.read(file_in)
seq.header="not a consensus"
seq.write(file_out)
file_out.close()
file_in.close()
os.system("cp "+filename+".oriented"+
" "+filename+
".malign.fa")
os.system("cp "+filename+".oriented"+
" "+filename+
".malign.fa.cons")
sys.exit(1)
seq_in_set=0
nb_consensus=0
count_set=0
set_size=size_db
while set_size>max_set_size:
set_size=set_size/2
tmp_file_out=open(tmp_consensus_filename,'w')
last_seq=0
while 1:
#read subset of sequence
seq.read(file_in)
if seq.sequence!=None:
if seq.getLength() < max_len and seq.getLength() > min_len:
seq.write(tmp_file_out)
seq_in_set=seq_in_set+1
else:
if seq.getLength() > max_len:
print seq.header+" too long!!"
if not seq.header.find(" too long, not aligned"):
seq.header=seq.header+" too long, not aligned"
seq.write(file_out)
if seq.getLength() < min_len:
print seq.header+" too short!!"
else:
last_seq=1
if seq_in_set==0:
return count_set
# aligne subset
if seq_in_set==set_size or last_seq:
count_set=count_set+1
print "aligning the set #",count_set," of ",seq_in_set," sequences"
tmp_file_out.close()
if seq_in_set>1:
os.system("nice malign "+tmp_consensus_filename
+" 20 -8 16 4 > "
+tmp_consensus_filename+".malign"
+str(count_set)+".fa")
os.system("nice consensusFastaAli.py -n "
+str(min_base_nb)+" "
+tmp_consensus_filename
+".malign"+str(count_set)+".fa ")
os.system("cp "+tmp_consensus_filename+
".malign"+str(count_set)+".fa "
+filename+
".malign"+str(count_set)+".fa")
else:
os.system("cp "+tmp_consensus_filename+
" "+filename+
".malign"+str(count_set)+".fa")
os.system("cp "+tmp_consensus_filename+
" "+tmp_consensus_filename+
".malign"+str(count_set)+".fa.cons")
os.system("cat "+tmp_consensus_filename+
".malign"+str(count_set)+\
".fa.cons >> "+consensus_filename)
seq_in_set=0
tmp_file_out=open(tmp_consensus_filename,'w')
if set_size==size_db or last_seq: break
tmp_file_out.close()
file_out.close()
file_in.close()
os.system("rm "+tmp_consensus_filename+"* "+filename+".oriented" )
return count_set
def dbRepShow(inFileName,reputer_param="-allmax -f -r -c -p -l 10"):
"""
deprecated
"""
import reputer
import Gnuplot
tmpname=os.tmpnam();
g = Gnuplot.Gnuplot(debug=1)
g('set data style lines')
g('set terminal postscript landscape color')
g('set output "'+inFileName+'.ps"')
allsum=0
n=0
max=0
min=0
file_db=open(inFileName)
seq=Bioseq()
numseq=0
while 1:
seq.read(file_db)
if seq.sequence==None:
break
n=n+1
replist=reputer.find(seq,reputer_param)
fileEmpty=open(tmpname+"E","w")
fileF=open(tmpname+"F","w")
fileR=open(tmpname+"R","w")
fileC=open(tmpname+"C","w")
fileP=open(tmpname+"P","w")
F=0
R=0
C=0
P=0
fileEmpty.write("%d\t%d"%(seq.getLength(),seq.getLength()))
for i in replist.rep_list:
if i.type=="F":
F=1
fileF.write(str(i.pos1)\
+"\t"+str(i.pos2)\
+"\n"+str(i.pos1+i.length1)\
+"\t"+\
str(i.pos2+i.length2)+"\n\n")
elif i.type=="R":
R=1
fileR.write(str(i.pos1)\
+"\t"+str(i.pos2)\
+"\n"+str(i.pos1+i.length1)\
+"\t"+\
str(i.pos2+i.length2)+"\n\n")
elif i.type=="C":
C=1
fileC.write(str(i.pos1)\
+"\t"+str(i.pos2)\
+"\n"+str(i.pos1+i.length1)\
+"\t"+\
str(i.pos2+i.length2)+"\n\n")
elif i.type=="P":
P=1
fileP.write(str(i.pos1)\
+"\t"+str(i.pos2)\
+"\n"+str(i.pos1+i.length1)\
+"\t"+\
str(i.pos2+i.length2)+"\n\n")
fileF.close()
fileR.close()
fileC.close()
fileP.close()
fileEmpty.close()
g.title(seq.header)
cmd='l='+str(seq.getLength())+\
'\nset xrange [0:l]\nset yrange [0:l]\nplot "'+\
tmpname+'E" notitle with dots'
if F:
cmd=cmd+', "'+tmpname+'F" title "F" with lines 1 1'
if R:
cmd=cmd+', "'+tmpname+'R" title "R" with lines 2 1'
if C:
cmd=cmd+', "'+tmpname+'C" title "C" with lines 3 1'
if P:
cmd=cmd+', "'+tmpname+'P" title "P" with lines 4 1'
g(cmd)
numseq=numseq+1
file_db.close()
os.system("gv "+inFileName+".ps")
os.system("rm "+tmpname+"*")
def filterClassifConsensus( inFileName, outFileName, filterSSRs, maxLengthToFilterSSRs, filterHostGenes, filterConfused, filterNoCat, nbAlignSeqNoCat, verbose=0 ): """ Filter each consensus according to the classification in its header. @param inFileName: name of the input fasta file @type inFileName: string @param outFileName: name of the output fasta file @type outFileName: string @param filterSSRs: filter the consensus classified as SSR @type filterSSRs: boolean @param maxLengthToFilterSSRs: length below which a SSR is filtered @type maxLengthToFilterSSRs: integer @param filterSSRs: filter the consensus classified as HostGene @type filterSSRs: boolean @param filterConfused: filter the consensus classified as confused @type filterConfused: boolean @param filterNoCat: filter the consensus classified as NoCat @type filterNoCat: boolean @param nbAlignSeqNoCat: minimum number of sequences in the MSA from which the NoCat consensus as been built @type nbAlignSeqNoCat: string @param verbose: verbosity level @type verbose: integer """ if outFileName == "": outFileName = "%s.filtered" % ( inFileName ) nbAlignSeqNoCat = int( nbAlignSeqNoCat ) if verbose > 0: print "input file: %s" % ( inFileName ) print "output file: %s" % ( outFileName ) if filterSSRs: if maxLengthToFilterSSRs == 0: print "filter SSRs" else: print "filter SSRs (<%ibp)" % ( maxLengthToFilterSSRs ) if filterHostGenes: print "filter host's genes" if filterNoCat: print "filter NoCat" if filterConfused: print "filter confused" sys.stdout.flush() inFile = open( inFileName, "r" ) outFile = open( outFileName, "w" ) bs = Bioseq() nbInSeq = 0 nbRmv = 0 while True: bs.read( inFile ) if bs.header == None: break nbInSeq += 1 if verbose > 1: print bs.header if filterSSRs == True and "SSR" in bs.header and ( maxLengthToFilterSSRs == 0 or bs.getLength() <= maxLengthToFilterSSRs ): nbRmv += 1 if verbose > 1: print "filtered !" elif filterHostGenes == True and "HostGene" in bs.header: nbRmv += 1 if verbose > 1: print "filtered !" elif filterConfused == True and "confusedness=yes" in bs.header: nbRmv += 1 if verbose > 1: print "filtered !" elif filterNoCat == True and "NoCat" in bs.header: algoMSA = "" for i in ["Map","MAP","Malign","Mafft","Prank","Clustalw","Muscle","Tcoffee"]: if i in bs.header: algoMSA = i regexp = ".*" + algoMSA + "_(\d*)\|.*" header = re.match(regexp, bs.header) nb = header.group(1) nbAlignSeq = int( nb ) if nbAlignSeq <= nbAlignSeqNoCat: nbRmv += 1 if verbose > 1: print "filtered !" else: bs.write( outFile ) else: bs.write( outFile ) inFile.close() outFile.close() if verbose > 0: print "nb of input seq: %i" % ( nbInSeq ) print "nb of filtered seq: %i" % ( nbRmv ) sys.stdout.flush()