def create_alnmt(start, end, chrom, name):
'''
Given 1-based blast start, end and chromosome, return an HTSeq Alignment object
'''
iv = create_genomic_interval(start, end, chrom)
r = HTSeq.SequenceWithQualities(b"", str(name), b"")
alnmt = HTSeq.Alignment(r, iv)
return alnmt
counts_transcriptome_edit = [0]
ref_rlen = -1
for almnt in almnt_file:
# Constructing the reference alignment.
total_rlen = len(almnt.read)
if ref_rlen < 0:
ref_rlen = total_rlen
counts_genome += [0] * total_rlen
counts_transcriptome += [0] * total_rlen
counts_transcriptome_edit += [0] * total_rlen
elif ref_rlen != total_rlen:
raise RuntimeError, "differing read lengths are not supported"
ref_pos = int(almnt.read.name)
s_almnt = HTSeq.Alignment(almnt.read, HTSeq.GenomicInterval(cur_chr, ref_pos, ref_pos+total_rlen, "+"))
source_ids = set()
for iv, val in source_features[s_almnt.iv].steps():
source_ids |= val
# Checking whether it's actually aligned to the new genome
if not almnt.aligned or almnt.aQual < args.minqual:
counts_genome[0] += 1
if len(source_ids):
counts_transcriptome[0] += 1
counts_transcriptome_edit[0] += 1
continue
# Computing the number of matching bases (edit distance includes indels, so we add them back in)
edit_distance = almnt.optional_field("NM")
total_match =- edit_distance
