for achrom in regionDict:
print "%s: processing %d regions" % (achrom, len(regionDict[achrom]))
for region in regionDict[achrom]:
border = 0
if usePeaks:
(rstart, rstop, rlen, peakPos, peakHeight) = region
border = 200
else:
(rstart, rstop, rlen) = region
if rlen > maxsize:
print "%s:%d-%d length %d > %d max region size - skipping" % (
achrom, rstart, rstop, rlen, maxsize)
continue
try:
seq = hg.sequence(achrom, rstart - border, rlen + 2 * border)
except:
print "problem with %s" % str((rstart, rstop, rlen))
continue
if usePeaks:
topPos = peakPos - rstart
if peakHeight > minHitThresh:
ncregions[achrom].append(
(rstart, rstop, rlen, [topPos], peakHeight))
index += 1
elif doDataset:
thechrom = 'chr' + achrom
print '.'
hitDict = hitRDS.getReadsDict(chrom=thechrom,
withWeight=True,
doMulti=True,
for (start, stop, length) in regions[chrom]:
regionList.append((chrom, start, length))
if usePeak:
regionList.sort()
regionList.reverse()
notFoundIndex = 0
currentChrom = ''
count = 0
for tuple in regionList:
if usePeak:
(rpeakheight, rchrom, start, length, rpeakpos) = tuple
else:
(rchrom, start, length) = tuple
try:
seq = hg.sequence(rchrom, start, length)
except:
print "couldn't retrieve %s %d %d - skipping" % (rchrom, start, length)
continue
count += 1
numHits = -1
if usePeak:
peakpos = rpeakpos
if useRank:
numHits = count
else:
numHits = rpeakheight
elif doRDS:
if rchrom != currentChrom:
fullchrom = 'chr' + rchrom
hitDict = hitRDS.getReadsDict(chrom=fullchrom)
def main(argv):
if len(argv) < 4:
print 'usage: python %s genome gtf outfilename [-polyA length]' % argv[0]
sys.exit(1)
genome = argv[1]
gtf=argv[2]
outputfilename = argv[3]
doPolyA=False
if '-polyA' in argv:
doPolyA=True
tailsize=int(argv[argv.index('-polyA')+1])
tail=''
for i in range(tailsize):
tail=tail+'A'
print 'will add a polyA tail of ', tailsize, 'nt'
outfile = open(outputfilename, 'w')
hg = Genome(genome)
j=0
lineslist = open(gtf)
TranscriptDict={}
for line in lineslist:
j+=1
if j % 100000 == 0:
print j, 'lines processed'
if line.startswith('#'):
continue
fields=line.strip().split('\t')
if fields[2]!='exon':
continue
if 'transcript_name "' in fields[8]:
TranscriptID=fields[8].split('transcript_name "')[1].split('";')[0]
else:
TranscriptID=fields[8].split('transcript_id "')[1].split('";')[0]
if TranscriptDict.has_key(TranscriptID):
pass
else:
TranscriptDict[TranscriptID]=[]
chr=fields[0]
left=int(fields[3])
right=int(fields[4])
orientation=fields[6]
TranscriptDict[TranscriptID].append((chr,left,right,orientation))
g=0
print 'Found', len(TranscriptDict.keys()), 'transcripts'
for transcript in TranscriptDict.keys():
g+=1
if g % 1000 == 0:
print g, 'transcripts sequences processed'
sequence=''
leftEnds=[]
rightEnds=[]
TranscriptDict[transcript].sort()
orientation = TranscriptDict[transcript][0][3]
if orientation=='+' or orientation=='F':
for (chr,left,right,orientation) in TranscriptDict[transcript]:
leftEnds.append(left)
rightEnds.append(right)
try:
sequence=sequence+hg.sequence(chr[3:len(chr)],left,right-left)
print "can't retrieve sequence"
except:
for p in range(left,right-left):
try:
sequence=sequence+hg.sequence(chr[3:len(chr)],p,1)
except:
sequence=sequence+'N'
missed+=1
sense='plus_strand'
if orientation=='-' or orientation=='R':
for (chr,left,right,orientation) in reversed(TranscriptDict[transcript]):
leftEnds.append(left)
rightEnds.append(right)
try:
exonsequence=hg.sequence(chr[3:len(chr)],left-1,right-left+1)
sequence=sequence+getReverseComplement(exonsequence)
except:
for p in range(left-1,right-left+1):
try:
sequence=sequence+getReverseComplement(hg.sequence(chr[3:len(chr)],p,1))
except:
sequence=sequence+'N'
missed+=1
sense='minus_strand'
LeftEnd=min(leftEnds)
RightEnd=max(rightEnds)
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense
outfile.write(outline+'\n')
if doPolyA:
outfile.write(sequence+tail+'\n')
else:
outfile.write(sequence+'\n')
outfile.close()
def main(argv):
if len(argv) < 4:
print 'usage: python %s genome gtf outfilename [-polyA length]' % argv[
0]
sys.exit(1)
genome = argv[1]
gtf = argv[2]
outputfilename = argv[3]
doPolyA = False
if '-polyA' in argv:
doPolyA = True
tailsize = int(argv[argv.index('-polyA') + 1])
tail = ''
for i in range(tailsize):
tail = tail + 'A'
print 'will add a polyA tail of ', tailsize, 'nt'
outfile = open(outputfilename, 'w')
hg = Genome(genome)
j = 0
lineslist = open(gtf)
TranscriptDict = {}
for line in lineslist:
j += 1
if j % 100000 == 0:
print j, 'lines processed'
if line.startswith('#'):
continue
fields = line.strip().split('\t')
if fields[2] != 'exon':
continue
if 'transcript_name "' in fields[8]:
TranscriptID = fields[8].split('transcript_name "')[1].split(
'";')[0]
else:
TranscriptID = fields[8].split('transcript_id "')[1].split('";')[0]
if TranscriptDict.has_key(TranscriptID):
pass
else:
TranscriptDict[TranscriptID] = []
chr = fields[0]
left = int(fields[3])
right = int(fields[4])
orientation = fields[6]
TranscriptDict[TranscriptID].append((chr, left, right, orientation))
g = 0
print 'Found', len(TranscriptDict.keys()), 'transcripts'
for transcript in TranscriptDict.keys():
g += 1
if g % 1000 == 0:
print g, 'transcripts sequences processed'
sequence = ''
leftEnds = []
rightEnds = []
TranscriptDict[transcript].sort()
orientation = TranscriptDict[transcript][0][3]
if orientation == '+' or orientation == 'F':
for (chr, left, right, orientation) in TranscriptDict[transcript]:
leftEnds.append(left)
rightEnds.append(right)
try:
sequence = sequence + hg.sequence(chr[3:len(chr)], left,
right - left)
print "can't retrieve sequence"
except:
for p in range(left, right - left):
try:
sequence = sequence + hg.sequence(
chr[3:len(chr)], p, 1)
except:
sequence = sequence + 'N'
missed += 1
sense = 'plus_strand'
if orientation == '-' or orientation == 'R':
for (chr, left, right,
orientation) in reversed(TranscriptDict[transcript]):
leftEnds.append(left)
rightEnds.append(right)
try:
exonsequence = hg.sequence(chr[3:len(chr)], left - 1,
right - left + 1)
sequence = sequence + getReverseComplement(exonsequence)
except:
for p in range(left - 1, right - left + 1):
try:
sequence = sequence + getReverseComplement(
hg.sequence(chr[3:len(chr)], p, 1))
except:
sequence = sequence + 'N'
missed += 1
sense = 'minus_strand'
LeftEnd = min(leftEnds)
RightEnd = max(rightEnds)
outline = '>' + transcript + ':' + chr + ':' + str(
LeftEnd) + '-' + str(RightEnd) + '-' + sense
outfile.write(outline + '\n')
if doPolyA:
outfile.write(sequence + tail + '\n')
else:
outfile.write(sequence + '\n')
outfile.close()
regionstart = exonStops[index] - maxBorder
alreadySeen[chrom].append((exonStops[index], exonStarts[index + 1]))
beforeLen = exonLengths[index]
afterLen = exonLengths[index + 1]
if (beforeLen + afterLen) < maxBorder + spacer:
#print 'splice chr%s:%d-%d too short: %d' % (chrom, exonStops[index], exonStarts[index + 1], beforeLen + afterLen)
missedCount += 1
continue
if (beforeLen + afterLen) < 2 * maxBorder:
depressedCount += 1
if beforeLen > maxBorder:
beforeLen = maxBorder
if afterLen > maxBorder:
afterLen = maxBorder
try:
beforeSplice = hg.sequence(chrom, exonStops[index] - maxBorder,
maxBorder)
afterSplice = hg.sequence(chrom, exonStarts[index + 1], maxBorder)
#beforeSplice = hg.sequence(chrom, exonStops[index] - beforeLen, beforeLen)
#afterSplice = hg.sequence(chrom, exonStarts[index + 1], afterLen)
except:
if doVerbose:
print "could not get chr%s:%d-%d" % (chrom, exonStops[index],
exonStarts[index + 1])
continue
outstring = '>%s%s%d%s%d\n%s\n' % (
name, delimiter, index, delimiter, regionstart,
spacerseq + beforeSplice.upper() + afterSplice.upper() + spacerseq)
outfile.write(outstring)
splicefileindex += 1
spliceCounter += 1
if spliceCounter > 10000:
countList = []
posList = []
index = 0
regionList = []
for rchrom in regions:
if 'rand' in rchrom or 'M' in rchrom or 'hap' in rchrom:
continue
for (start, stop, length) in regions[rchrom]:
regionList.append((rchrom, start, length))
notFoundIndex = 0
currentChrom = ''
for (rchrom, start, length) in regionList:
seq = hg.sequence(rchrom, start, length)
if doDataset:
if rchrom != currentChrom:
fullchrom = 'chr' + rchrom
hitDict = hitRDS.getReadsDict(chrom=fullchrom, withWeight=True, doMulti=True)
currentChrom = rchrom
(topPos, numHits, smoothArray, numPlus) = findPeak(hitDict[rchrom], start, length, doWeight=True)
if len(topPos) == 0:
print 'topPos error'
peakpos = topPos[0]
peakscore = smoothArray[peakpos]
if peakscore == 0.:
peakscore = -1.
if normalize:
numHits /= normalizeBy
peakscore /= normalizeBy
from cistematic.genomes import Genome
print '%s: version 1.1' % sys.argv[0]
if len(sys.argv) < 5:
print 'usage: python %s genome merlen chrAny:start-stop outfile' % sys.argv[
0]
sys.exit(1)
genome = sys.argv[1]
merlen = int(sys.argv[2])
location = sys.argv[3]
outfilename = sys.argv[4]
(chrom, pos) = location.split(':')
chrom = chrom[3:]
(start, stop) = pos.split('-')
start = int(start)
regionlength = int(stop) - start + 1
hg = Genome(genome)
seq = hg.sequence(chrom, start, regionlength)
outfile = open(outfilename, 'w')
print 'writing %d %d-mers' % (regionlength - merlen, merlen)
for index in range(regionlength - merlen):
outfile.write(seq[index:index + merlen].upper() + '\n')
outfile.close()
def main(argv):
if len(argv) < 3:
print 'usage: python %s genome gtf outfilename [-spliced] [-class_code symbol]' % argv[0]
print ' this script will output the translation of all three possible reading frames; stop codons will be converted to a .'
sys.exit(1)
genome = argv[1]
gtf=argv[2]
outputfilename = argv[3]
doSpliced=False
if '-spliced' in argv:
doSpliced=True
print 'will only look at transciprs with more than one exon'
doClassCode=False
if '-class_code' in argv:
doClassCode=True
class_code=argv[argv.index('-class_code')+1]
print 'will only look at transciprs if class code', class_code
CodonDict={'GCU':'A', 'GCC':'A', 'GCA':'A', 'GCG':'A',
'UUA':'L', 'UUG':'L', 'CUU':'L', 'CUC':'L', 'CUA':'L', 'CUG':'L',
'CGU':'R', 'CGC':'R', 'CGA':'R', 'CGG':'R', 'AGA':'R', 'AGG':'R',
'AAA':'K', 'AAG':'K',
'AAU':'N', 'AAC':'N',
'AUG':'M',
'GAU':'D', 'GAC':'D',
'UUU':'F', 'UUC':'F',
'UGU':'C', 'UGC':'C',
'CCU':'P', 'CCC':'P', 'CCA':'P', 'CCG':'P',
'CAA':'Q', 'CAG':'Q',
'UCU':'S', 'UCC':'S', 'UCA':'S', 'UCG':'S', 'AGU':'S', 'AGC':'S',
'GAA':'E', 'GAG':'E',
'ACU':'T', 'ACC':'T', 'ACA':'T', 'ACG':'T',
'GGU':'G', 'GGC':'G', 'GGA':'G', 'GGG':'G',
'UGG':'W',
'CAU':'H', 'CAC':'H',
'UAU':'Y', 'UAC':'Y',
'AUU':'I', 'AUC':'I', 'AUA':'I',
'GUU':'V', 'GUC':'V', 'GUA':'V', 'GUG':'V',
'START':'AUG',
'UAA':'.',
'UGA':'.',
'UAG':'.'}
outfile = open(outputfilename, 'w')
hg = Genome(genome)
j=0
lineslist = open(gtf)
TranscriptDict={}
for line in lineslist:
j+=1
if j % 100000 == 0:
print j, 'lines processed'
if line.startswith('#'):
continue
fields=line.strip().split('\t')
if fields[2]!='exon':
continue
if doClassCode:
if 'class_code "' in fields[8]:
cc = fields[8].split('class_code "')[1].split('";')[0]
if cc != class_code:
continue
else:
continue
if 'transcript_name "' in fields[8]:
TranscriptID=fields[8].split('transcript_name "')[1].split('";')[0]
else:
TranscriptID=fields[8].split('transcript_id "')[1].split('";')[0]
if TranscriptDict.has_key(TranscriptID):
pass
else:
TranscriptDict[TranscriptID]=[]
chr=fields[0]
left=int(fields[3])
right=int(fields[4])
orientation=fields[6]
TranscriptDict[TranscriptID].append((chr,left,right,orientation))
g=0
print 'Found', len(TranscriptDict.keys()), 'transcripts'
for transcript in TranscriptDict.keys():
g+=1
if g % 1000 == 0:
print g, 'transcripts sequences processed'
TranscriptDict[transcript] = list(Set(TranscriptDict[transcript]))
if doSpliced:
if len(TranscriptDict[transcript]) == 1:
del TranscriptDict[transcript]
continue
sequence=''
leftEnds=[]
rightEnds=[]
orientation = TranscriptDict[transcript][0][3]
TranscriptDict[transcript].sort()
if orientation=='+':
for (chr,left,right,orientation) in TranscriptDict[transcript]:
try:
sequence=sequence+hg.sequence(chr[3:len(chr)],left,right-left)
except:
print "can't retrieve sequence", chr,left,right,orientation
for p in range(left,right-left):
try:
sequence=sequence+hg.sequence(chr[3:len(chr)],p,1)
except:
sequence=sequence+'N'
missed+=1
sense='plus_strand'
if orientation=='-':
for (chr,left,right,orientation) in reversed(TranscriptDict[transcript]):
try:
exonsequence=hg.sequence(chr[3:len(chr)],left-1,right-left+1)
sequence=sequence+getReverseComplement(exonsequence)
except:
print "can not retrieve sequence", chr,left,right,orientation
for p in range(left-1,right-left+1):
try:
sequence=sequence+getReverseComplement(hg.sequence(chr[3:len(chr)],p,1))
except:
sequence=sequence+'N'
missed+=1
sense='minus_strand'
if orientation=='.':
for (chr,left,right,orientation) in TranscriptDict[transcript]:
try:
sequence=sequence+hg.sequence(chr[3:len(chr)],left,right-left)
except:
print "can not retrieve sequence", chr,left,right,orientation
for p in range(left,right-left):
try:
sequence=sequence+hg.sequence(chr[3:len(chr)],p,1)
except:
sequence=sequence+'N'
missed+=1
sense='unknown_strand'
LeftEnd=TranscriptDict[transcript][0][1]
RightEnd=TranscriptDict[transcript][-1][2]
if orientation == '+' or orientation == '-':
sequence = sequence.upper().replace('T','U')
max_protein_length = len(sequence)
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame1'
outfile.write(outline+'\n')
protein = ''
for i in range(0,max_protein_length-3,3):
if 'N' in sequence[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence[i:i+3]]
outfile.write(protein+'\n')
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame2'
outfile.write(outline+'\n')
protein = ''
for i in range(1,max_protein_length-4,3):
if 'N' in sequence[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence[i:i+3]]
outfile.write(protein+'\n')
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame3'
outfile.write(outline+'\n')
protein = ''
for i in range(2,max_protein_length-5,3):
if 'N' in sequence[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence[i:i+3]]
outfile.write(protein+'\n')
else:
sequence1 = sequence.upper().replace('T','U')
sequence2 = getReverseComplement(sequence).upper().replace('T','U')
max_protein_length = len(sequence1)
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame1'
outfile.write(outline+'\n')
protein = ''
for i in range(0,max_protein_length-3,3):
if 'N' in sequence1[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence1[i:i+3]]
outfile.write(protein+'\n')
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame2'
outfile.write(outline+'\n')
protein = ''
for i in range(1,max_protein_length-4,3):
if 'N' in sequence1[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence1[i:i+3]]
outfile.write(protein+'\n')
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame3'
outfile.write(outline+'\n')
protein = ''
for i in range(2,max_protein_length-5,3):
if 'N' in sequence1[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence1[i:i+3]]
outfile.write(protein+'\n')
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame4'
outfile.write(outline+'\n')
protein = ''
for i in range(0,max_protein_length-3,3):
if 'N' in sequence2[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence2[i:i+3]]
outfile.write(protein+'\n')
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame5'
outfile.write(outline+'\n')
protein = ''
for i in range(1,max_protein_length-4,3):
if 'N' in sequence2[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence2[i:i+3]]
outfile.write(protein+'\n')
outline='>'+transcript+':'+chr+':'+str(LeftEnd)+'-'+str(RightEnd)+'-'+sense+'::frame6'
outfile.write(outline+'\n')
protein = ''
for i in range(2,max_protein_length-5,3):
if 'N' in sequence2[i:i+3]:
protein = protein + '.'
else:
protein = protein + CodonDict[sequence2[i:i+3]]
outfile.write(protein+'\n')
outfile.close()