def write_insertion_repeat_dir(data,outRepeat):
insSeq = data['contigSeqGenomeDir'][data['leftBpContigCoord']:data['rightBpContigCoord']-1]
rmLines = read_rm_file(data['contigSeqFileNameRM'])
# figure if sine is on + or -
rmDir = figure_sine_strand(rmLines)
if rmDir == '-':
insSeq = genutils.revcomp(insSeq)
outS = genutils.add_breaks_to_line(insSeq)
outRepeat.write('>%s\n%s\n' % (data['siteID'],outS))
def get_contig_seq(data):
print_dictionary(data)
contigsFileName = data['originalContigsDir'] + '/' + 'combined.scaffolds.fa'
contigSeq = genutils.read_fasta_file_to_list(contigsFileName)
cSeq = contigSeq[data['contigName']]['seq']
if data['contigDir'] == '-':
cSeq = genutils.revcomp(cSeq)
data['contigSeqFileName'] = data['alignOutDir'] + '/' + 'Contig.genomedir.fa'
data['contigSeqGenomeDir'] = cSeq
contigSeqStr = genutils.add_breaks_to_line(cSeq)
outFile = open(data['contigSeqFileName'],'w')
outFile.write('>contig\n%s\n' % contigSeqStr)
outFile.close()
contigSeq = genutils.read_fasta_to_string(data['contigSeqFileName'])
data['contigLen'] = len(data['contigSeqGenomeDir'])
def check_seq(fq,myData):
minScore = 49.0
result = {}
result['passChecks'] = False
# do not know the alignment orientation, so check both and take best score
alignRes = pairwise2.align.globalms(myData['linkerSeq'], fq['seq'], 2, -1, -.5, -.2,penalize_end_gaps=False)
alignResLinkerRC = pairwise2.align.globalms(myData['linkerSeqRC'], fq['seq'], 2, -1, -.5, -.2,penalize_end_gaps=False)
# compare scores
if alignRes[0][2] >= alignResLinkerRC[0][2]:
ls = myData['linkerSeq']
else:
ls = myData['linkerSeqRC']
alignRes = alignResLinkerRC
result['align'] = alignRes
# should only be one alignment. Otherwise
if len(alignRes) != 1:
# print 'have mulitple potential alignments'
# print fq['seq']
result['passChecks'] = False
return result
# check score
if alignRes[0][2] < minScore:
result['passChecks'] = False
return result
#figure out coordinates
# go through keeping track of pos to go form column number to bp number
# do the sequence in 1 based coordinates
seq1ColToPos = []
current = 0
for i in range(len(alignRes[0][0])): #[1,2,3,4], looking at alignment
if alignRes[0][0][i] != '-':
current += 1
seq1ColToPos.append(current)
seq2ColToPos = []
current = 0
for i in range(len(alignRes[0][1])):
if alignRes[0][1][i] != '-':
current += 1
seq2ColToPos.append(current)
linkerColStart = -1
linkerColEnd = -1
for i in range(len(seq1ColToPos)):
if seq1ColToPos[i] == 1 and linkerColStart == -1:
linkerColStart = i
if seq1ColToPos[i] == len(myData['linkerSeq']) and linkerColEnd == -1:
linkerColEnd = i
# extract sequences -1 because python 0 based, colToSeq 1 based,
leftSeq = fq['seq'][0:seq2ColToPos[linkerColStart]-1]
linkerSeq = fq['seq'][seq2ColToPos[linkerColStart]-1:seq2ColToPos[linkerColEnd]]
rightSeq = fq['seq'][seq2ColToPos[linkerColEnd]:]
result['passChecks'] = True
# passess, so take out the sequence and quals
leftSeqQual = fq['qual33Str'][0:seq2ColToPos[linkerColStart]-1]
rightSeqQual = fq['qual33Str'][seq2ColToPos[linkerColEnd]:]
# need to reverse comp R1 due to structure of the library
leftSeq = genutils.revcomp(leftSeq)
leftSeqQual = leftSeqQual[::-1]
result['seq1'] = leftSeq
result['seq1Qual'] = leftSeqQual
result['seq2'] = rightSeq
result['seq2Qual'] = rightSeqQual
return result
###############################################################################
if options.outDir[-1] != '/':
options.outDir += '/'
# setup file location info
myData = {}
myData['filesToDelete'] = []
myData['filesToGzip'] = []
myData['r1fq'] = options.r1fq
myData['r2fq'] = options.r2fq
myData['sampleName'] = options.sampleName
myData['outDir'] = options.outDir
myData['linkerSeq'] = 'CTGCTGTACCGTTCTCCGTACAGCAG'
# rev of linker is also possible, since do not know orietnation
myData['linkerSeqRC'] = genutils.revcomp(myData['linkerSeq'])
print 'Processing %s' % myData['sampleName']
#run pear to join together reads that overlap
run_pear(myData)
count_num_not_assembled(myData)
print '%i reads were not assembled' % myData['numNotAssem']
count_num_discarded(myData)
print '%i reads were discarded' % myData['numDiscarded']
process_assembled(myData)
print '%i reads were assembled' % myData['numAssembled']
print '%i assembled reads failed the check' % myData['numFail']
print '%i assembled reads that passed the check but failed the length test (< 22bp)' % myData['lenFail']
print '%i total reads in original fastq' % myData['totReads']
outReads.write(nl)
inFile.close()
print 'Writing formated output read sequences to',outSeqName
inFile = open(samFile,'r')
for line in inFile:
line = line.rstrip()
line = line.split('\t')
samRec = genutils.parse_sam_line(line)
seq = samRec['seq']
qual = samRec['qual']
if samRec['reverseStrand'] is True:
qual = qual[::-1]
seq = genutils.revcomp(seq)
if samRec['isFirst'] is True:
i = 0
else:
i = 1
nl = [samRec['seqName'],sN,str(i),seq,qual]
nl = '\t'.join(nl) + '\n'
outSeq.write(nl)
inFile.close()
outSeq.close()
outReads.close()
print 'Now doing window bed to get reads associated with each call'
print 'Output to',outReadsNameIntersect