def get_transcripts_at_interval(chrom, start, end, strand, exon_trees):
features = []
for hit in exon_trees[chrom].find(start, end):
txlist = hit.value
for g,exon_num in txlist:
features.extend(g for g in txlist if cmp_strand(g.strand, strand))
return features
def get_transcripts_at_interval(chrom, start, end, strand, exon_trees):
features = []
for hit in exon_trees[chrom].find(start, end):
txlist = hit.value
for g, exon_num in txlist:
features.extend(g for g in txlist if cmp_strand(g.strand, strand))
return features
def get_genes_at_interval(chrom, start, end, strand, trees):
features = []
for hit in trees[chrom].find(start, end):
txlist = hit.value
# check for compatibility with overlapping genes
# TODO: debug strand-specific libraries such that
# antisense reads are not counted as part of overlapping
# sense genes. this should work but needs testing
features.extend(g for g in txlist if cmp_strand(g.strand, strand))
return features
def get_genes_at_interval(chrom, start, end, strand, trees):
features = []
for hit in trees[chrom].find(start, end):
txlist = hit.value
# check for compatibility with overlapping genes
# TODO: debug strand-specific libraries such that
# antisense reads are not counted as part of overlapping
# sense genes. this should work but needs testing
features.extend(g for g in txlist if cmp_strand(g.strand, strand))
return features
def get_transcripts_at_interval(chrom, start, end, strand, exon_intervals, exon_trees):
txsegs = []
for hit in exon_trees[chrom].find(start, end):
txlist = hit.value
# check for compatibility with overlapping genes
for g,exon_num in txlist:
g_exon_start, g_exon_end = g.exons[exon_num]
#print 'ALN', g.gene_name, g.tx_name, 'enum', exon_num, g.strand, strand, g_exon_start, g_exon_end, start, end
# strand must be compatible
if not cmp_strand(g.strand, strand):
continue
# get exon coordinates
g_exon_start, g_exon_end = g.exons[exon_num]
e_start = max(start, g_exon_start) - g_exon_start
e_end = min(end, g_exon_end) - g_exon_start
e_start_overhang = max(0, g_exon_start - start)
e_end_overhang = max(0, end - g_exon_end)
# get transcript coordinates
tx_start = sum((end-start) for start,end in g.exons[:exon_num])
tx_start = tx_start + e_start
tx_end = tx_start + (end - start)
# convert to negative strand if necessary
if g.strand == NEG_STRAND:
tx_length = sum((end - start) for start,end in g.exons)
tx_end, tx_start = (tx_length - tx_start, tx_length - tx_end)
exon_num = len(g.exons) - exon_num
txsegs.append(TranscriptAlignment(g=g,
exon_num=exon_num,
chrom=chrom,
start=start,
end=end,
e_start=e_start,
e_end=e_end,
e_start_overhang=e_start_overhang,
e_end_overhang=e_end_overhang,
tx_start=tx_start,
tx_end=tx_end))
return txsegs
