def get_features_within_interval_patch(contig,
beg,
end,
quicksect_obj,
fname_corr=None):
"""
Query a quicksect object for features overlapping
the given interval on a chromosome
"""
all_overlaps = []
for feature_type, feature_quicksect in quicksect_obj.iteritems():
if fname_corr:
fname = fname_corr[feature_type]
else:
fname = feature_type
overlaps = feature_quicksect.find(
quicksect.Feature(beg, end, chr=contig))
if not overlaps:
continue
for e in overlaps:
all_overlaps.append(
(fname, e.info["gene_id"], str(e.info["start"]),
str(e.info["end"]), e.info[fname + "_number"],
e.info["source"], e.info["strand"], e.info))
return all_overlaps
def make_composite(tr_list):
if not tr_list:
return 0, 0, 0
features = []
feature_sizes = {'cds': 0, '3utr': 0, '5utr': 0}
starts, stops, strands, names = range(4)
all_breaks = []
for tr in tr_list:
exon, cds, utr5, utr3 = make_transcript(tr, what='model')
all_breaks.extend(cds[starts] + utr5[starts] + utr3[starts] +
cds[stops] + utr5[stops] + utr3[stops])
#print cds, utr5, utr3
#for (start, stop, strand, name) in zip(zip(*cds) + zip(*utr5) + zip(*utr3)):
#interval = interval.insert(quicksect.Feature(start, stop, strand, name))
try:
features += [
quicksect.Feature(start, stop - 1, strand, name)
for (start, stop, strand,
name) in zip(*cds) + zip(*utr5) + zip(*utr3)
]
except Exception, e:
sys.stderr.write('Problem during packing @ {}: {}\n'.format(
tr, str(e)))
sys.stderr.write('exons: {}\n'.format(exon))
sys.stderr.write('CDSs: {}\n'.format(cds))
sys.stderr.write('5UTRs: {}\n'.format(utr5))
sys.stderr.write('3UTRs: {}\n'.format(utr3))
def get_features_within_interval(contig, beg, end, quicksect_obj):
"""
Query a quicksect object for features overlapping
the given interval on a chromosome
"""
all_overlaps = []
for feature_type, feature_quicksect in quicksect_obj.iteritems():
overlaps = feature_quicksect.find(
quicksect.Feature(beg, end, chr=contig))
if not overlaps:
continue
for e in overlaps:
all_overlaps.append(
(feature_type, e.info["gene_id"], str(e.info["start"]),
str(e.info["end"]), e.info["source"], e.info["strand"],
e.info["attr"]))
return all_overlaps
def create_quicksect(loci, feature_types):
"""
Create quicksect dictionary for looking up only exons from each single locus
"""
quicksect_obj = {
feature_type: quicksect.IntervalTree()
for feature_type in feature_types
}
for locus in loci:
for feature_type in feature_types:
for feature in loci[locus].features[feature_type]:
quicksect_obj[feature_type].insert(
quicksect.Feature(feature.iv.start,
feature.iv.end - 1,
chr=feature.iv.chrom,
info=getInfo(feature)))
return quicksect_obj
def create_quicksect_patch(loci, feature_types):
"""
Create quicksect dictionary for looking up only exons from each single locus
"""
quicksect_obj = {
feature_type: quicksect.IntervalTree()
for feature_type in feature_types
}
for locus in loci:
for feature_type in feature_types:
for feature in loci[locus]['features'][feature_type]:
quicksect_obj[feature_type].insert(
quicksect.Feature(feature['start'],
feature['end'] - 1,
chr=feature['chrom'],
info=feature))
return quicksect_obj