def extractSeq(feature, blastDb, dx, dy):
"""Extract the translated sequence of a feature and DNA
sequence of the surrounding the genomic region.
@param feature: Feature object. Mandatory attributes: accession, sStart, sEnd.
@param blastDb: Blast database.
@param dx: Length of sequence to extract upstream.
@param dy: Length of sequence to extract downstream.
@returns: tuple of fasta strings (DNA, protein).
"""
# Extract hmmer hit & translate
header,seq = fasta.getSequence(blastDb, feature.accession,
start=feature.sStart, end=feature.sEnd, strand=feature.strand)
protein = '\n'.join(['>' + header, sequence.translate(seq)])
if feature.strand=='+':
start = feature.sStart-dx
end = feature.sEnd+dy
else:
start = feature.sStart-dy
end = feature.sEnd+dx
if start<0 or end>5000000:
raise Exception('Out of block bounds.')
# Extract surrounding DNA sequence
header,seq = fasta.getSequence(blastDb, feature.accession,
start=start, end=end)
header = '%(accession)s:%(sStart)s-%(sEnd)s' % feature.__dict__
dna = '\n'.join(['>' + header, seq])
return dna, protein
for i, domain in enumerate(domains):
if i > 99: break
domain.domain = 'DEFB_%0.2i' % (i + 1)
domain.toGenomic(relative=True)
domain.addField('lowScoring', 'N')
domain.addField('pseudogene', 'N')
domain.addField('nCysteines', 0)
summary = []
h, s = fasta.getSequence(blastdb,
domain.accession,
start=domain.sStart,
end=domain.sEnd,
strand=domain.strand)
pep = sequence.translate(s)
if i > 59: domain.lowScoring = 'Y'
if '*' in pep:
domain.pseudogene = 'Y'
summary.append('Contains stops')
for aa in pep:
if aa == 'C':
domain.nCysteines += 1
if domain.nCysteines != 6:
summary.append('Has %i cysteines' % domain.nCysteines)
print >> summaryFile, '%s\t%s' % (domain.domain, '; '.join(summary))
if domain.pseudogene == 'Y' or domain.nCysteines < 5 or domain.nCysteines > 7:
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:
gStart,gEnd = gEnd,gStart
s = dna[gStart-1:gEnd]
s = sequence.reverseComplement(s)
print gStart, gEnd, len(s), len(s) % 3==0
print sequence.codons(s, remainder=True)
print sequence.translate(s)
print
writer.close()
from aaiter import find_orfs
from fastamasta import FastaReader
from sequence import translate, rev_complement
if __name__ == "__main__":
reader = FastaReader("data/12.dat")
seq = reader.readnext()[1]
rcseq = rev_complement(seq)
orfs = find_orfs(seq)
rorfs = find_orfs(rcseq)
candidates = set()
for orf in orfs:
candidates.add(translate(seq[orf[0]:orf[1]]))
for rorf in rorfs:
candidates.add(translate(rcseq[rorf[0]:rorf[1]]))
for candidate in candidates:
print candidate
domains = hmmer.loadDomains('DEFB.txt', seqType='BlockSixFrame')
print >> genomicFile, '\t'.join(domains[0].fields
+ ['strand', 'lowScoring', 'pseudogene', 'nCysteines'])
for i,domain in enumerate(domains):
if i>99: break
domain.domain = 'DEFB_%0.2i' % (i+1)
domain.toGenomic(relative=True)
domain.addField('lowScoring', 'N')
domain.addField('pseudogene', 'N')
domain.addField('nCysteines', 0)
summary = []
h,s = fasta.getSequence(blastdb, domain.accession,
start=domain.sStart, end=domain.sEnd, strand=domain.strand)
pep = sequence.translate(s)
if i>59: domain.lowScoring = 'Y'
if '*' in pep:
domain.pseudogene = 'Y'
summary.append('Contains stops')
for aa in pep:
if aa=='C':
domain.nCysteines += 1
if domain.nCysteines!=6:
summary.append('Has %i cysteines' % domain.nCysteines)
print >> summaryFile, '%s\t%s' % (domain.domain, '; '.join(summary))
if domain.pseudogene=='Y' or domain.nCysteines<5 or domain.nCysteines>7:
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])
if s!=orf: print h
writer.write(h,s + '\n')
writer.close()
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:
gStart, gEnd = gEnd, gStart
s = dna[gStart - 1:gEnd]
s = sequence.reverseComplement(s)
print gStart, gEnd, len(s), len(s) % 3 == 0
print sequence.codons(s, remainder=True)
print sequence.translate(s)
print
writer.close()
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:
gStart,gEnd = gEnd,gStart
dna = sequence.reverseComplement(s[gStart-1:gEnd])
print len(dna), len(dna) % 3==0
print sequence.codons(dna, remainder=True)
print sequence.translate(dna)
print
else:
domains = hmmer4.load_domains('hmmer/ORFs.txt')
for d in domains:
o = hmmer4.parseOrfHeader(d.accession)
print d
print o
gStart,gEnd = hmmer4.convertOrfToGenomic(d.sStart,d.sEnd,o.strand,o.start)
if o.strand=='+':