def Initialize():
header,seq = fasta.load('MHC_hg18.fa')
sixFrameIter = sequence.sixFrameTranslationIter(seq)
writer = fasta.MfaWriter('6frames.fa')
for frame,p in sixFrameIter:
print 'Frame:', frame
writer.write('%s:%i' % (header,frame),p)
writer.close()
sys.exit()
def Initialize():
header, seq = fasta.load('MHC_hg18.fa')
sixFrameIter = sequence.sixFrameTranslationIter(seq)
writer = fasta.MfaWriter('6frames.fa')
for frame, p in sixFrameIter:
print 'Frame:', frame
writer.write('%s:%i' % (header, frame), p)
writer.close()
sys.exit()
def loadCdsSeq(codingSeqFnL):
'''Given a list NCBI coding sequence file names (fna) load the sequences and store in a dictionary keyed by locus tag.
'''
seqD={}
for fn in codingSeqFnL:
for header,seq in fasta.load(fn):
locusTag = header.split('locus_tag=')[1].split(']')[0]
seqD[locusTag]=seq
return seqD
def loadProt(protFnL):
'''Given a list of file names of fasta files with the gene name as
header, load the sequences and store in a dictionary keyed by protein
name.
'''
seqD={}
for fn in protFnL:
for header,seq in fasta.load(fn):
gn = header.split()[0][1:]
seqD[gn]=seq
return seqD
def loadProt(protFnL):
'''Given a list of file names of fasta files with the gene name as
header, load the sequences and store in a dictionary keyed by protein
name.
'''
seqD = {}
for fn in protFnL:
for header, seq in fasta.load(fn):
gn = header.split()[0][1:]
seqD[gn] = seq
return seqD
def createGeneSpeciesMap(protFile):
f=open(protFile,'r')
while True:
s=f.readline()
if s=='':
break
s=s.rstrip('\n')
L=s.rstrip().split('/')
k=L[2].split('.')[0]
print k,
strainInfo=fasta.load(s)
for gene in strainInfo:
q=gene[0]
q=q[1:]
q=q.split()
q=q[0]
print q,
print
return
except:
print """
Your input files cannot be found.
"""
sys.exit(2)
# Read in the cd-hit clustered file
try:
cluster_list = cluster_file.read()
except:
print 'Cannot open', cluster_file, '\n'
sys.exit(2)
# Read in the FASTA file and check to make sure it's in FASTA format
try:
fasta_dict_raw = fasta.load(fasta_file)
except:
print '\n', sys.argv[2], 'does not appear to be a fasta file\n'
sys.exit(2)
fasta_dict = {}
# Just take everything before the first space in the first line of the FASTA file as
# the key. This is also how cd-hit takes the name, so the keys will match.
for fasta_key in fasta_dict_raw:
new_fasta = fasta_key.split(' ')
new_key = new_fasta[0]
fasta_dict[new_key] = fasta_dict_raw[fasta_key]
usage = "%prog [options] <gff file> <fasta reference sequence>"
parser = OptionParser(usage=usage)
parser.add_option("-o", "--output", dest="oFilename",
help="Output filename", default=None)
parser.add_option("-f", "--features", dest="features",
help="Features to extract", default=['exon'])
options, args = parser.parse_args(sys.argv)
if len(args)!=3:
sys.exit(__doc__)
gffFilename = args[1]
faFilename = args[2]
data = gff.load(gffFilename)
header,seq = fasta.load(faFilename)
if options.oFilename:
oFile = open(options.oFilename, 'w')
else:
oFile = sys.stdout
writer = fasta.MfaWriter(oFile)
for name in data:
s = []
extrema = []
for f in data[name]:
if f.type in options.features:
if f.strand=='+':
start,end = f.start,f.end
_seq = seq[start-1:end]
action="store_true",
dest="complement",
help="Complement sequence",
default=False)
parser.add_option(
"-b", "--reverseComplement", "--revComp",
action="store_true",
dest="reverseComplement",
help="Reverse complement sequence", default=False)
options, args = parser.parse_args(sys.argv)
iFilename = args[1]
start = int(args[2])
end = int(args[3])
header,seq = fasta.load(iFilename)
s = seq[start-1:end]
h = '%s %i-%i' % (header,start,end)
if options.reverse:
s = sequence.reverse(s)
h += '(r)'
elif options.complement:
s = sequence.complement(s)
h += '(c)'
elif options.reverseComplement:
s = sequence.reverse_complement(s)
h += '(rc)'
fasta.pretty(h, s, width=options.width)
if __name__ == "__main__":
if len(sys.argv) !=4:
sys.stderr.write("""
Usage: python seedAlign.fa protDB.fa outfile.txt
""")
sys.exit(-1)
seedAlignFN = sys.argv[1]
dbSeqFN = sys.argv[2]
outFN = sys.argv[3]
# create profile HMM based on seed alignment
seedAlignL = [seq for header,seq in fasta.load(seedAlignFN)]
# load db
dbL = fasta.load(dbSeqFN)
outL=[]
model=hmmmodel.HmmModel(seedAlignFN)
for hd,sseq in dbL:
# load db
print("aligning: "+hd)
alignment= hmmAlign.HmmAlign(sseq,model)
score=alignment.subtractShuffleMean()
outL.append((score,hd))
#!/usr/bin/python
# chrom.py
import sys
import fasta
assembly_file = sys.argv[1]
contigs_file = sys.argv[2]
contigs = {}
for f in fasta.load(open(contigs_file)):
contigs[f.name] = f
chrom = "chrI"
print ">" + chrom
for line in open(assembly_file):
fields = line.strip().split()
s, e = int(fields[1]), int(fields[2])
l = e - s
if fields[3][0:3] == "gap":
sys.stdout.write("N" * l)
else:
strand = fields[5]
if strand == "+":
sys.stdout.write(contigs[fields[3]].seq)
elif strand == "-":
sys.stdout.write(contigs[fields[3]].reverse_complement().seq)
elif strand == ".":
sys.stdout.write(contigs[fields[3]].seq)
sys.exit(2) # Check to make sure the input file exists and can be opened try: fasta_file = open(filename) except: print "This file could not be opened" sys.exit(2) # Check to make sure the input file is in FASTA format try: fasta_data = fasta.load(fasta_file) except: print 'This file does not seem to be a fasta file. Please try again with a fasta file' sys.exit(0) # Write out the FASTA input file to a new file, because CD-HIT doesn't handle # all input file types correctly new_fasta_file = dirname+'/tmp/input_fasta_file.fa' try: input_fasta_file = open(new_fasta_file, 'wt') except: print 'Cannot open', new_fasta_file, 'for writing a temporary fasta file'
#!/usr/bin/env python
"""
testHmmer.py
Author: Tony Papenfuss
Date: Fri Sep 1 09:09:02 EST 2006
"""
import os, sys, re
import hmmer4, fasta, sequence
h,s = fasta.load('seq/HLA-A.fa')
L = len(s)
if False:
domains = hmmer4.load_domains('hmmer/6frames.txt')
for d in domains:
p = hmmer4.parseSixFrameHeader(d.accession)
print d
print p.name, p.frame
gStart,gEnd,strand = hmmer4.convert6FrameToGenomic(d.sStart,d.sEnd,p.frame,L)
print gStart,gEnd,strand
if strand=='+':
dna = s[gStart-1:gEnd]
print len(dna), len(dna) % 3==0
print sequence.codons(dna, remainder=True)
print sequence.translate(dna)
else:
except:
print """
Your input files cannot be found.
"""
sys.exit(2)
# Read in the cd-hit clustered file
try:
cluster_list = cluster_file.read()
except:
print 'Cannot open', cluster_file, '\n'
sys.exit(2)
# Read in the FASTA file and check to make sure it's in FASTA format
try:
fasta_dict_raw = fasta.load(fasta_file)
except:
print '\n', sys.argv[2], 'does not appear to be a fasta file\n'
sys.exit(2)
fasta_dict = {}
# Just take everything before the first space in the first line of the FASTA file as
# the key. This is also how cd-hit takes the name, so the keys will match.
for fasta_key in fasta_dict_raw:
new_fasta = fasta_key.split(' ')
new_key = new_fasta[0]
fasta_dict[new_key] = fasta_dict_raw[fasta_key]
# Output file for the list of all the sequences in each cluster
#!/usr/bin/env python
"""
extractORFs.py
Author: Tony Papenfuss
Date: Wed Aug 23 08:52:58 EST 2006
"""
import os, sys
import re, copy
import fasta, sequence, hmmer3
seqFilename = sys.argv[1]
header, seq = fasta.load(seqFilename)
header = header.split()[0]
L = len(seq)
pattern = re.compile('\*')
minLen = 10
sixFrameIter = sequence.sixFrameTranslationIter(seq)
writer = fasta.MfaWriter(sys.stdout)
i = 0
for frame, p in sixFrameIter:
print >> sys.stderr, 'Frame:', frame
matchIter = pattern.finditer(p)
match = matchIter.next()
def Initialize():
header,seq = fasta.load('MHC_hg18.fa')
sixFrameIter = sequence.sixFrameTranslationIter(seq)
writer = fasta.MfaWriter('6frames.fa')
for frame,p in sixFrameIter:
print 'Frame:', frame
writer.write('%s:%i' % (header,frame),p)
writer.close()
sys.exit()
# Initialize()
header,seq = fasta.load('MHC_hg18.fa')
L = len(seq)
hstart = header.split()[0]
pattern = re.compile('\*|X{200,}')
minLen = 20
# sixFrameIter = sequence.sixFrameTranslationIter(seq)
sixFrameIter = fasta.load_iter('6frames.fa')
writer = fasta.MfaWriter('ORFs.fa')
i = 0
for h,p in sixFrameIter:
hmmerFrame = int(h.split(':')[-1])
frame = hmmer.hmmer2frame[hmmerFrame]
print >> sys.stderr, 'Frame:', frame
def Initialize():
header, seq = fasta.load('MHC_hg18.fa')
sixFrameIter = sequence.sixFrameTranslationIter(seq)
writer = fasta.MfaWriter('6frames.fa')
for frame, p in sixFrameIter:
print 'Frame:', frame
writer.write('%s:%i' % (header, frame), p)
writer.close()
sys.exit()
# Initialize()
header, seq = fasta.load('MHC_hg18.fa')
L = len(seq)
hstart = header.split()[0]
pattern = re.compile('\*|X{200,}')
minLen = 20
# sixFrameIter = sequence.sixFrameTranslationIter(seq)
sixFrameIter = fasta.load_iter('6frames.fa')
writer = fasta.MfaWriter('ORFs.fa')
i = 0
for h, p in sixFrameIter:
hmmerFrame = int(h.split(':')[-1])
frame = hmmer.hmmer2frame[hmmerFrame]
print >> sys.stderr, 'Frame:', frame
f = open(aabrhAlignmentFN, "w")
randStr = str(random.randrange(1e5))
intempAlignFN = "/tmp/tempAlign" + randStr + ".fa"
outtempAlignFN = "/tmp/tempAlign" + randStr + ".afa"
for orthos in aabrhL:
tempf = open(intempAlignFN, "w")
writeSeqBlock(tempf, orthos, seqD)
tempf.close()
# align the temp file
os.system("muscle -in " + intempAlignFN + " -out " + outtempAlignFN)
# load aligned file into a sequence dict
alSeqL = fasta.load(outtempAlignFN)
alSeqD = {}
for hd, sq in alSeqL:
alSeqD[hd[1:]] = sq
# write aligned fasta block into main output file, in the
# original order (muscle messes up this order).
for gene in orthos:
commonName, locusTag, descrip, chrom, start, end, strand = geneInfoD[
gene]
f.write(">" + gene + " " + locusTag + "\n")
f.write(alSeqD[gene] + "\n")
f.write("\n")
# delete the temp files
os.system("rm " + intempAlignFN)
"""
extractORFs.py
Author: Tony Papenfuss
Date: Wed Aug 23 08:52:58 EST 2006
"""
import os, sys
import re, copy
import fasta, sequence, hmmer3
seqFilename = sys.argv[1]
header,seq = fasta.load(seqFilename)
header = header.split()[0]
L = len(seq)
pattern = re.compile('\*')
minLen = 10
sixFrameIter = sequence.sixFrameTranslationIter(seq)
writer = fasta.MfaWriter(sys.stdout)
i = 0
for frame,p in sixFrameIter:
print >> sys.stderr, 'Frame:', frame
matchIter = pattern.finditer(p)
match = matchIter.next()
Date: Wed Aug 23 08:52:58 EST 2006
"""
import os, sys
import re, copy
import fasta, sequence
pattern = re.compile('[\*|X{200,}]')
minLen = 20
i = 0
writer = fasta.MfaWriter('ORFs.fa')
filename = sys.argv[1]
header, dna = fasta.load(filename)
header = header.strip()
orfIter = sequence.extractOrfsIter(dna, minLen=minLen, pattern=pattern)
for i, gStart, gEnd, orf in orfIter:
h = '%s.%i.%i-%i Length=%i' % (header, i, gStart, gEnd, len(orf))
writer.write(h, orf)
fasta.pretty(h, orf)
if gStart < gEnd:
s = dna[gStart - 1:gEnd]
print gStart, gEnd, len(s), len(s) % 3 == 0
print sequence.codons(s, remainder=True)
print sequence.translate(s)
else:
#!/usr/bin/env python
"""
orfTest.py
Author: Tony Papenfuss
Date: Tue Aug 22 20:14:57 EST 2006
"""
import os, sys
import fasta, sequence
header,seq = fasta.load('NKC.fa')
orfIterator = fasta.load_iter('ORFs.fa')
writer = fasta.MfaWriter('ORFs2.fa')
for h,orf in orfIterator:
chrom,block,orfId,limits = h.split()[0].split('.')
start,end = limits.split('-')
start = int(start)
end = int(end)
if start>end:
strand = '-'
start,end = end,start
s = sequence.translate(sequence.reverseComplement(seq[start-1:end]))
else:
strand = '+'
s = sequence.translate(seq[start-1:end])
help="Complement sequence",
default=False)
parser.add_option("-b",
"--reverseComplement",
"--revComp",
action="store_true",
dest="reverseComplement",
help="Reverse complement sequence",
default=False)
options, args = parser.parse_args(sys.argv)
iFilename = args[1]
start = int(args[2])
end = int(args[3])
header, seq = fasta.load(iFilename)
s = seq[start - 1:end]
h = '%s %i-%i' % (header, start, end)
if options.reverse:
s = sequence.reverse(s)
h += '(r)'
elif options.complement:
s = sequence.complement(s)
h += '(c)'
elif options.reverseComplement:
s = sequence.reverse_complement(s)
h += '(rc)'
fasta.pretty(h, s, width=options.width)
#!/usr/bin/python
# Usage: chromByLen.py 454LargeContigs.fna chrI
import fasta
import sys
gap = "N" * 100
contig_file, cn, outdir = sys.argv[1], sys.argv[2], sys.argv[3]
# Load fasta file w/ contigs sequences. Typically, 454LargeContigs.fna
contigs = dict([(x.name, x) for x in fasta.load(open(sys.argv[1]))])
# get sorted list of contigs by length
lengths = [(contigs[x].length, x) for x in contigs]
lengths.sort(reverse=True)
cn = sys.argv[2]
# output BED tracks
cf = open(outdir + '/contigs.bed', 'w')
gf = open(outdir + '/gaps.bed', 'w')
cstart = 0
gn = 1
for length, name in lengths:
print >> cf, cn, cstart, cstart+length, name, 1000, '+'
cstart = cstart + length
print >> gf, cn, cstart, cstart+100, 'gap%s' % gn
cstart = cstart + len(gap)
sys.exit(2) # Check to make sure the input file exists and can be opened try: fasta_file = open(filename) except: print "This file could not be opened" sys.exit(2) # Check to make sure the input file is in FASTA format try: fasta_data = fasta.load(fasta_file) except: print 'This file does not seem to be a fasta file. Please try again with a fasta file' sys.exit(0) # Write out the FASTA input file to a new file, because CD-HIT doesn't handle # all input file types correctly new_fasta_file = dirname+'/tmp/input_fasta_file.fa' try: input_fasta_file = open(new_fasta_file, 'wt') except: print 'Cannot open', new_fasta_file, 'for writing a temporary fasta file'
"""
import os, sys
import re, copy
import fasta, sequence
pattern = re.compile('[\*|X{200,}]')
minLen = 20
i = 0
writer = fasta.MfaWriter('ORFs.fa')
filename = sys.argv[1]
header,dna = fasta.load(filename)
header = header.strip()
orfIter = sequence.extractOrfsIter(dna, minLen=minLen, pattern=pattern)
for i,gStart,gEnd,orf in orfIter:
h = '%s.%i.%i-%i Length=%i' % (header,i,gStart,gEnd,len(orf))
writer.write(h, orf)
fasta.pretty(h, orf)
if gStart<gEnd:
s = dna[gStart-1:gEnd]
print gStart, gEnd, len(s), len(s) % 3==0
print sequence.codons(s, remainder=True)
print sequence.translate(s)
else:
#!/usr/bin/env python
"""codingscore2.py <fasta file> <ghmm cfg> <gff file>
In windows of step evaluates log(P(S|q = coding) / P(S|q = noncoding))
and outputs to file
"""
from common.GHMM import *
from common.feature import *
import fasta, sys
if len(sys.argv) != 4:
print 'Invalid arguments'
print __doc__
sys.exit(0)
head,seq = fasta.load(sys.argv[1])
ghmm = GHMM(sys.argv[2])
features = Features(sys.argv[3])
sequence = sequenceDict(seq, 5)
outfile = open(sys.argv[1] + '.scr', 'w')
posscorecnc = []
posscore = []
for ref in features:
pos = 0
for generef in features[ref]:
gene = features[ref][generef]
genecoords = gene.coords
for (start, end) in genecoords:
coding1 = max(ghmm.content['eintn'].probEmit(sequence, start, end+1, 0, '+'), ghmm.content['eintn'].probEmit(sequence, start, end+1, 1, '+'), ghmm.content['eintn'].probEmit(sequence, start, end+1, 2, '+'))
f=open(aabrhAlignmentFN,"w")
randStr = str(random.randrange(1e5))
intempAlignFN="/tmp/tempAlign"+randStr+".fa"
outtempAlignFN="/tmp/tempAlign"+randStr+".afa"
for orthos in aabrhL:
tempf = open(intempAlignFN,"w")
writeSeqBlock(tempf,orthos,seqD)
tempf.close()
# align the temp file
os.system("muscle -in "+ intempAlignFN + " -out " + outtempAlignFN)
# load aligned file into a sequence dict
alSeqL=fasta.load(outtempAlignFN)
alSeqD={}
for hd,sq in alSeqL:
alSeqD[hd[1:]]=sq
# write aligned fasta block into main output file, in the
# original order (muscle messes up this order).
for gene in orthos:
commonName,locusTag,descrip,chrom,start,end,strand=geneInfoD[gene]
f.write(">"+gene+" "+locusTag+"\n")
f.write(alSeqD[gene]+"\n")
f.write("\n")
# delete the temp files
os.system("rm "+intempAlignFN)
os.system("rm "+outtempAlignFN)
#!/usr/bin/python
# Usage: chromByLen.py 454LargeContigs.fna chrI
import fasta
import sys
gap = "N" * 100
contig_file, cn, outdir = sys.argv[1], sys.argv[2], sys.argv[3]
# Load fasta file w/ contigs sequences. Typically, 454LargeContigs.fna
contigs = dict([(x.name, x) for x in fasta.load(open(sys.argv[1]))])
# get sorted list of contigs by length
lengths = [(contigs[x].length, x) for x in contigs]
lengths.sort(reverse=True)
cn = sys.argv[2]
# output BED tracks
cf = open(outdir + '/contigs.bed', 'w')
gf = open(outdir + '/gaps.bed', 'w')
cstart = 0
gn = 1
for length, name in lengths:
print >> cf, cn, cstart, cstart + length, name, 1000, '+'
cstart = cstart + length
print >> gf, cn, cstart, cstart + 100, 'gap%s' % gn
cstart = cstart + len(gap)
gn += 1
#!/usr/bin/python
# chrom.py
import sys
import fasta
assembly_file = sys.argv[1]
contigs_file = sys.argv[2]
contigs = {}
for f in fasta.load(open(contigs_file)):
contigs[f.name] = f
chrom = 'chrI'
print '>' + chrom
for line in open(assembly_file):
fields = line.strip().split()
s, e = int(fields[1]), int(fields[2])
l = e - s
if fields[3][0:3] == 'gap':
sys.stdout.write('N' * l)
else:
strand = fields[5]
if strand == '+':
sys.stdout.write(contigs[fields[3]].seq)
elif strand == '-':
sys.stdout.write(contigs[fields[3]].reverse_complement().seq)
elif strand == '.':
sys.stdout.write(contigs[fields[3]].seq)
import sys, fasta
import hmmmodel
import hmmAlign
#### Main
if __name__ == "__main__":
if len(sys.argv) != 3:
sys.stderr.write("""
Usage: python seedAlign.fa protSeq.fa
""")
sys.exit(-1)
seedAlignFN = sys.argv[1]
protSeqFN = sys.argv[2]
# create profile HMM based on seed alignment
model = hmmmodel.HmmModel(seedAlignFN)
# load db
sseq = fasta.load(protSeqFN)[0][1]
alignment = hmmAlign.HmmAlign(sseq, model)
score = alignment.subtractShuffleMean()
print(score)
