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
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)
pepWriter = fasta.MfaWriter('DEFB_extracted_pep.fa')
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))
def getSequence(self, blastDb, padFivePrime=0, padThreePrime=0):
start = max(1,self.start-padFivePrime)
end = self.end+padThreePrime
h,s = fasta.getSequence(blastDb, self.reference, start, end, self.strand)
return h,s
def getSequence(self, blastDb, padFivePrime=0, padThreePrime=0):
start = max(1, self.start - padFivePrime)
end = self.end + padThreePrime
h, s = fasta.getSequence(blastDb, self.chrom, start, end, self.strand)
return h, s
