def _get_hits(fn):
hits = Counter()
with open(fn) as handle:
for line in handle:
attr = read_attributes(line, "=")
query_sequence = attr['TS'].replace("U", "T")
if query_sequence and query_sequence.find("N") > -1:
continue
idu = make_id(query_sequence)
hits[idu] += 1
return hits
def _get_hits(fn):
hits = Counter()
with open(fn) as handle:
for line in handle:
attr = read_attributes(line, "=")
query_sequence = attr['TS'].replace("U", "T")
if query_sequence and query_sequence.find("N") > -1:
continue
idu = make_id(query_sequence)
hits[idu] += 1
return hits
def read_file(fn, precursors, matures):
with open(fn) as inh:
for line in inh:
if line.startswith("#"):
continue
cols = line.strip().split("\t")
attr = read_attributes(line)
t5 = _get_5p(precursors[attr["Parent"]],
matures[attr["Parent"]][attr["Name"]],
attr["Variant"])
t3 = _get_3p(precursors[attr["Parent"]],
matures[attr["Parent"]][attr["Name"]],
attr["Variant"])
add = _get_add(attr["Read"], attr["Variant"])
print[attr["Variant"], t5, t3, add]
def _read_file(fn, precursors, matures, out_dir):
samples = read_samples(fn)
for sample in samples:
with open(os.path.join(out_dir, "%s.mirna" % sample), 'w') as outh:
print("\t".join(
["seq", "name", "freq", "mir", "start", "end",
"mism", "add", "t5", "t3", "s5", "s3", "DB",
"precursor", "ambiguity"]), file=outh)
with open(fn) as inh:
for line in inh:
if line.startswith("#"):
continue
cols = line.strip().split("\t")
attr = read_attributes(line)
read = read_id(attr["UID"])
t5 = variant_to_5p(precursors[attr["Parent"]],
matures[attr["Parent"]][attr["Name"]],
attr["Variant"])
t3 = variant_to_3p(precursors[attr["Parent"]],
matures[attr["Parent"]][attr["Name"]],
attr["Variant"])
add = variant_to_add(read,
attr["Variant"])
mature_sequence = get_mature_sequence(
precursors[attr["Parent"]],
matures[attr["Parent"]][attr["Name"]])
mm = align_from_variants(read,
mature_sequence,
attr["Variant"])
if len(mm) > 1:
continue
elif len(mm) == 1:
mm = "".join(map(str, mm[0]))
else:
mm = "0"
hit = attr["Hits"] if "Hits" in attr else "1"
logger.debug("exporter::isomir::decode %s" % [attr["Variant"],
t5, t3, add, mm])
# Error if attr["Read"] doesn't exist
line = [read, attr["Read"], "0", attr["Name"], cols[1], cols[2],
mm, add, t5, t3, "NA", "NA", "miRNA", attr["Parent"], hit]
for sample, counts in zip(samples, attr["Expression"].split(",")):
with open(os.path.join(out_dir, "%s.mirna" % sample),
'a') as outh:
line[2] = counts
print("\t".join(line), file=outh)
def _read_file(fn, precursors, matures, out_dir):
samples = read_samples(fn)
for sample in samples:
with open(os.path.join(out_dir, "%s.mirna" % sample), 'w') as outh:
print("\t".join([
"seq", "name", "freq", "mir", "start", "end", "mism", "add",
"t5", "t3", "s5", "s3", "DB", "precursor", "ambiguity"
]),
file=outh)
with open(fn) as inh:
for line in inh:
if line.startswith("#"):
continue
cols = line.strip().split("\t")
attr = read_attributes(line)
read = read_id(attr["UID"])
t5 = variant_to_5p(precursors[attr["Parent"]],
matures[attr["Parent"]][attr["Name"]],
attr["Variant"])
t3 = variant_to_3p(precursors[attr["Parent"]],
matures[attr["Parent"]][attr["Name"]],
attr["Variant"])
add = variant_to_add(read, attr["Variant"])
mature_sequence = get_mature_sequence(
precursors[attr["Parent"]],
matures[attr["Parent"]][attr["Name"]])
mm = align_from_variants(read, mature_sequence, attr["Variant"])
if len(mm) > 1:
continue
elif len(mm) == 1:
mm = "".join(map(str, mm[0]))
else:
mm = "0"
hit = attr["Hits"] if "Hits" in attr else "1"
logger.debug("exporter::isomir::decode %s" %
[attr["Variant"], t5, t3, add, mm])
# Error if attr["Read"] doesn't exist
line = [
read, attr["Read"], "0", attr["Name"], cols[1], cols[2], mm,
add, t5, t3, "NA", "NA", "miRNA", attr["Parent"], hit
]
for sample, counts in zip(samples, attr["Expression"].split(",")):
with open(os.path.join(out_dir, "%s.mirna" % sample),
'a') as outh:
line[2] = counts
print("\t".join(line), file=outh)
def _calc_stats(fn):
"""
Read files and parse into categories
"""
samples = _get_samples(fn)
lines = []
seen = set()
with open(fn) as inh:
for line in inh:
if line.startswith("#"):
continue
cols = line.strip().split("\t")
logger.debug("## STATS: attribute %s" % cols[8])
attr = read_attributes(line)
if "-".join([attr['Variant'], attr['Name']]) in seen:
continue
seen.add("-".join([attr['Variant'], attr['Name']]))
lines.extend(_classify(cols[2], attr, samples))
df = _summary(lines)
return df
def read_file(fn, args):
"""
Read isomiR-SEA file and convert to mirtop GFF format.
Args:
*fn(str)*: file name with isomiR-SEA output information.
*database(str)*: database name.
*args(namedtuple)*: arguments from command line.
See *mirtop.libs.parse.add_subparser_gff()*.
Returns:
*reads (nested dicts)*:gff_list has the format as
defined in *mirtop.gff.body.read()*.
"""
database = args.database
gtf = args.gtf
sep = " " if args.out_format == "gtf" else "="
map_mir = mapper.read_gtf_to_mirna(gtf)
reads = defaultdict(dict)
reads_in = 0
sample = os.path.splitext(os.path.basename(fn))[0]
hits = _get_hits(fn)
logger.debug("ISOMIRSEA::SAMPLE::%s" % sample)
with open(fn) as handle:
for line in handle:
cols = line.strip().split("\t")
attr = read_attributes(line, "=")
query_name = attr['TS']
query_sequence = attr['TS'].replace("U", "T")
start = int(cols[3])
end = int(cols[4])
isomirseq_iso = attr['ISO']
if query_name not in reads and query_sequence == None:
continue
if query_sequence and query_sequence.find("N") > -1:
continue
counts = attr["TC"]
chrom = cols[0]
# logger.debug("SEQBUSTER:: cigar {cigar}".format(**locals()))
cigar = attr['CI'].replace("U", "T")
idu = make_id(query_sequence)
isoformat = cigar2variants(cigar, query_sequence, attr['ISO'])
logger.debug("\nISOMIRSEA::NEW::query: {query_sequence}\n"
" precursor {chrom}\n"
" name: {query_name}\n"
" idu: {idu}\n"
" start: {start}\n"
" cigar: {cigar}\n"
" iso: {isoformat}\n"
" variant: {isoformat}".format(**locals()))
source = "isomiR" if isoformat != "NA" else "ref_miRNA"
strand = "+"
database = cols[1]
mirName = attr['MIN'].split()[0]
preName = attr['PIN'].split()[0]
score = "."
Filter = attr['FILTER']
isotag = attr['ISO']
tchrom, tstart = _genomic2transcript(map_mir[mirName], chrom,
start)
start = start if not tstart else tstart
chrom = chrom if not tstart else tchrom
end = start + len(query_sequence)
hit = hits[idu]
fields = {
'seq_name': query_sequence,
'idseq': idu,
'name': mirName,
'parent': preName,
'variant': isoformat,
'cigar': cigar,
'counts': counts,
'filter': Filter,
'hits': hit,
'chrom': chrom,
'start': start,
'end': end,
'database': database,
'source': source,
'score': score,
'strand': strand
}
# TODO: convert to genomic if args.out_genomic
line = feature(fields).line
if args.add_extra:
extra = variant_with_nt(line, args.precursors, args.matures)
line = "%s Changes %s;" % (line, extra)
line = paste_columns(feature(line), sep=sep)
if start not in reads[chrom]:
reads[chrom][start] = []
if Filter == "Pass":
reads_in += 1
reads[chrom][start].append([idu, chrom, counts, sample, line])
logger.info("Hits: %s" % reads_in)
return reads
def read_file(fn, database, gtf):
"""
read bam file and perform realignment of hits
"""
map_mir = mapper.read_gtf_to_mirna(gtf)
reads = defaultdict(dict)
reads_in = 0
sample = os.path.splitext(os.path.basename(gtf))[0]
hits = _get_hits(fn)
with open(fn) as handle:
for line in handle:
cols = line.strip().split("\t")
attr = read_attributes(line, "=")
query_name = attr['TS']
query_sequence = attr['TS'].replace("U", "T")
start = int(cols[3])
end = int(cols[4])
isomirseq_iso = attr['ISO']
if query_name not in reads and query_sequence == None:
continue
if query_sequence and query_sequence.find("N") > -1:
continue
counts = attr["TC"]
chrom = cols[0]
# logger.debug("SEQBUSTER:: cigar {cigar}".format(**locals()))
cigar = attr['CI'].replace("U", "T")
idu = make_id(query_sequence)
isoformat = cigar2variants(cigar, query_sequence, attr['ISO'])
logger.debug("\nSOMIRSEA::NEW::query: {query_sequence}\n"
" precursor {chrom}\n"
" name: {query_name}\n"
" idu: {idu}\n"
" start: {start}\n"
" cigar: {cigar}\n"
" iso: {isoformat}\n"
" variant: {isoformat}".format(**locals()))
source = "isomiR" if isoformat != "NA" else "ref_miRNA"
strand = "+"
database = cols[1]
mirName = attr['MIN'].split()[0]
preName = attr['PIN'].split()[0]
score = "."
Filter = attr['FILTER']
isotag = attr['ISO']
tchrom, tstart = genomic2transcript(map_mir[mirName], chrom, start)
start = start if not tstart else tstart
chrom = chrom if not tstart else tchrom
end = start + len(query_sequence)
hit = hits[idu]
attrb = (
"Read {query_sequence}; UID {idu}; Name {mirName}; Parent {preName}; Variant {isoformat}; Isocode {isotag}; Cigar {cigar}; Expression {counts}; Filter {Filter}; Hits {hit};"
).format(**locals())
res = (
"{chrom}\t{database}\t{source}\t{start}\t{end}\t{score}\t{strand}\t.\t{attrb}"
).format(**locals())
if start not in reads[chrom]:
reads[chrom][start] = []
if Filter == "Pass":
reads_in += 1
reads[chrom][start].append([idu, chrom, counts, sample, res])
logger.info("Hits: %s" % reads_in)
return reads
def read_file(fn, args):
"""
Read isomiR-SEA file and convert to mirtop GFF format.
Args:
*fn(str)*: file name with isomiR-SEA output information.
*database(str)*: database name.
*args(namedtuple)*: arguments from command line.
See *mirtop.libs.parse.add_subparser_gff()*.
Returns:
*reads (nested dicts)*:gff_list has the format as
defined in *mirtop.gff.body.read()*.
"""
database = args.database
gtf = args.gtf
sep = " " if args.out_format == "gtf" else "="
map_mir = mapper.read_gtf_to_mirna(gtf)
reads = defaultdict(dict)
reads_in = 0
sample = os.path.splitext(os.path.basename(fn))[0]
hits = _get_hits(fn)
logger.debug("ISOMIRSEA::SAMPLE::%s" % sample)
with open(fn) as handle:
for line in handle:
cols = line.strip().split("\t")
attr = read_attributes(line, "=")
query_name = attr['TS']
query_sequence = attr['TS'].replace("U", "T")
start = int(cols[3])
end = int(cols[4])
isomirseq_iso = attr['ISO']
if query_name not in reads and query_sequence == None:
continue
if query_sequence and query_sequence.find("N") > -1:
continue
counts = attr["TC"]
chrom = cols[0]
# logger.debug("SEQBUSTER:: cigar {cigar}".format(**locals()))
cigar = attr['CI'].replace("U", "T")
idu = make_id(query_sequence)
isoformat = cigar2variants(cigar, query_sequence, attr['ISO'])
logger.debug("\nISOMIRSEA::NEW::query: {query_sequence}\n"
" precursor {chrom}\n"
" name: {query_name}\n"
" idu: {idu}\n"
" start: {start}\n"
" cigar: {cigar}\n"
" iso: {isoformat}\n"
" variant: {isoformat}".format(**locals()))
source = "isomiR" if isoformat != "NA" else "ref_miRNA"
strand = "+"
database = cols[1]
mirName = attr['MIN'].split()[0]
preName = attr['PIN'].split()[0]
score = "."
Filter = attr['FILTER']
isotag = attr['ISO']
tchrom, tstart = _genomic2transcript(map_mir[mirName],
chrom, start)
start = start if not tstart else tstart
chrom = chrom if not tstart else tchrom
end = start + len(query_sequence)
hit = hits[idu]
attrb = ("Read {query_sequence}; UID {idu}; Name {mirName};"
" Parent {preName}; Variant {isoformat};"
" Isocode {isotag}; Cigar {cigar}; Expression {counts};"
" Filter {Filter}; Hits {hit};").format(**locals())
line = ("{chrom}\t{database}\t{source}\t{start}\t{end}\t"
"{score}\t{strand}\t.\t{attrb}").format(**locals())
if args.add_extra:
extra = variant_with_nt(line, args.precursors, args.matures)
line = "%s Changes %s;" % (line, extra)
line = paste_columns(read_gff_line(line), sep=sep)
if start not in reads[chrom]:
reads[chrom][start] = []
if Filter == "Pass":
reads_in += 1
reads[chrom][start].append([idu, chrom, counts, sample, line])
logger.info("Hits: %s" % reads_in)
return reads