def reader(args):
"""
Realign BAM hits to miRBAse to get better accuracy and annotation
"""
if args.low_memory:
read.reader(args)
return None
samples = []
database = mapper.guess_database(args)
args.database = database
precursors = fasta.read_precursor(args.hairpin, args.sps)
args.precursors = precursors
matures = mapper.read_gtf_to_precursor(args.gtf)
args.matures = matures
# TODO check numbers of miRNA and precursors read
# TODO print message if numbers mismatch
out_dts = dict()
if args.keep_name and len(args.files) > 1:
logger.warning("--keep-name when running multiple samples\n"
"can generate wrong results if the\n"
"name read is different across sample\n"
"for the same sequence.")
for fn in args.files:
fn = op.normpath(fn)
if args.format != "gff":
sample = op.splitext(op.basename(fn))[0]
samples.append(sample)
fn_out = op.join(args.out, sample + ".%s" % args.out_format)
if args.format == "BAM":
reads = _read_bam(fn, args)
elif args.format == "seqbuster":
reads = seqbuster.read_file(fn, args)
elif args.format == "srnabench":
out_dts[fn] = srnabench.read_file(fn, args)
elif args.format == "prost":
reads = prost.read_file(fn, precursors, database, args.gtf)
elif args.format == "isomirsea":
out_dts[fn] = isomirsea.read_file(fn, args)
elif args.format == "manatee":
out_dts[fn] = manatee.read_file(fn, database, args)
elif args.format == "optimir":
out_dts[fn] = optimir.read_file(fn, args)
elif args.format == "gff":
samples.extend(header.read_samples(fn))
out_dts[fn] = body.read(fn, args)
continue
if args.format not in ["isomirsea", "srnabench", "manatee", 'optimir']:
ann = annotate(reads, matures, precursors)
out_dts[fn] = body.create(ann, database, sample, args)
h = header.create([sample], database, header.make_tools(args.format))
_write(out_dts[fn], h, fn_out, args)
# merge all reads for all samples into one dict
if args.low_memory:
return None
merged = merge.merge(out_dts, samples)
fn_merged_out = op.join(args.out, "mirtop.%s" % args.out_format)
_write(merged,
header.create(samples, database, header.make_tools([args.format])),
fn_merged_out, args)
def reader(args):
"""
Realign BAM hits to miRBAse to get better accuracy and annotation
"""
samples = []
database = mapper.guess_database(args.gtf)
args.database = database
precursors = fasta.read_precursor(args.hairpin, args.sps)
args.precursors = precursors
matures = mapper.read_gtf_to_precursor(args.gtf)
args.matures = matures
# TODO check numbers of miRNA and precursors read
# TODO print message if numbers mismatch
out_dts = dict()
for fn in args.files:
if args.format != "gff":
sample = op.splitext(op.basename(fn))[0]
samples.append(sample)
fn_out = op.join(args.out, sample + ".%s" % args.out_format)
if args.format == "BAM":
reads = _read_bam(fn, args)
elif args.format == "seqbuster":
reads = seqbuster.read_file(fn, args)
elif args.format == "srnabench":
out_dts[fn] = srnabench.read_file(fn, args)
elif args.format == "prost":
reads = prost.read_file(fn, precursors, database, args.gtf)
elif args.format == "isomirsea":
out_dts[fn] = isomirsea.read_file(fn, args)
elif args.format == "gff":
samples.extend(header.read_samples(fn))
out_dts[fn] = body.read(fn, args)
continue
if args.format not in ["isomirsea", "srnabench"]:
ann = annotate(reads, matures, precursors)
out_dts[fn] = body.create(ann, database, sample, args)
h = header.create([sample], database, "")
_write(out_dts[fn], h, fn_out)
# merge all reads for all samples into one dict
merged = merge.merge(out_dts, samples)
fn_merged_out = op.join(args.out, "mirtop.%s" % args.out_format)
_write(merged, header.create(samples, database, ""), fn_merged_out)
def test_collapse(self):
"""testing GFF function"""
from mirtop.libs import logger
from mirtop.mirna import mapper, fasta
from mirtop.gff import body, header
logger.initialize_logger("test", True, True)
logger = logger.getLogger(__name__)
precursors = fasta.read_precursor("data/examples/annotate/hairpin.fa",
"hsa")
# depend on https://github.com/miRTop/mirtop/issues/6
matures = mapper.read_gtf_to_precursor(
"data/examples/annotate/hsa.gff3")
# matures = mirtop.mirna.read_mature("data/examples/annotate/mirnas.gff", "hsa")
from mirtop.bam import bam
bam_fn = "data/aligments/collapsing-isomirs.sam"
reads = bam.read_bam(bam_fn, precursors)
ann = bam.annotate(reads, matures, precursors)
fn = bam_fn + ".gff"
h = header.create(bam_fn, ["example"], "miRBase21")
gff = body.create(ann, "miRBase21", "example", fn, header)
print gff
return True
def reader(args):
"""
Realign BAM hits to miRBase to get better accuracy and annotation
"""
samples = []
database = mapper.guess_database(args)
args.database = database
precursors = fasta.read_precursor(args.hairpin, args.sps)
args.precursors = precursors
matures = mapper.read_gtf_to_precursor(args.gtf)
args.matures = matures
# TODO check numbers of miRNA and precursors read
# TODO print message if numbers mismatch
if args.keep_name and len(args.files) > 1:
logger.warning("--keep-name when running multiple samples\n"
"can generate wrong results if the\n"
"name read is different across sample\n"
"for the same sequence.")
for fn in args.files:
fn = op.normpath(fn)
if args.format != "gff":
sample = op.splitext(op.basename(fn))[0]
samples.append(sample)
fn_out = op.join(args.out, sample + ".%s" % args.out_format)
h = header.create([sample], args.database, "")
out_handle = open(fn_out, 'w')
print(h, file=out_handle)
if args.format == "BAM":
if args.genomic:
low_memory_genomic_bam(fn, sample, out_handle, args)
else:
low_memory_bam(fn, sample, out_handle, args)
elif args.format == "seqbuster":
seqbuster.read_file_low_memory(fn, sample, args, out_handle)
else:
raise ValueError("%s not supported for low memory" % args.format)
out_handle.close()
def reader(args):
"""
Realign BAM hits to miRBAse to get better accuracy and annotation
"""
database = mapper.guess_database(args.gtf)
# hairpin, mirna = download_mirbase(args)
precursors = fasta.read_precursor(args.hairpin, args.sps)
matures = mapper.read_gtf_to_precursor(args.gtf)
# check numnbers of miRNA and precursors read
# print message if numbers mismatch
out_dts = dict()
for fn in args.files:
sample = op.splitext(op.basename(fn))[0]
fn_out = op.join(args.out, sample + ".gff")
if args.format == "BAM":
reads = _read_bam(fn, precursors)
elif args.format == "seqbuster":
reads = seqbuster.read_file(fn, precursors)
custom = seqbuster.header()
elif args.format == "srnabench":
reads = srnabench.read_gile(fn, precursors)
h = header.create([sample], database, "")
ann = annotate(reads, matures, precursors)
out_dts[fn] = body.create(ann, database, sample, fn_out, h)
default=False)
parser.add_option("-p", "--prefix", help="output name")
parser.add_option("--seed",
help="set up seed for reproducibility.",
default=None)
(options, args) = parser.parse_args()
if options.seed:
random.seed(options.seed)
full_fq = "%s_full.fq" % options.prefix
clean_fq = "%s_clean.fq" % options.prefix
out_gff = "%s.gff" % options.prefix
if os.path.exists(full_fq):
os.remove(full_fq)
if os.path.exists(clean_fq):
os.remove(clean_fq)
pre = fasta.read_precursor(options.fa, "")
mir = mapper.read_gtf_to_precursor(options.gtf)
nt = ['A', 'T', 'G', 'C']
gffs = dict()
h = header.create(["sampleX"], "miRBase1", "")
for precursor in pre:
seq = pre[precursor]
gffs.update(create_iso(precursor, mir, seq, options.numsim, options.exp))
_write(gffs, h, out_gff)
help="give expression", default=False)
parser.add_option("-p", "--prefix", help="output name")
parser.add_option("--seed", help="set up seed for reproducibility.", default = None)
(options, args) = parser.parse_args()
if options.seed:
random.seed(options.seed)
full_fq = "%s_full.fq" % options.prefix
clean_fq = "%s_clean.fq" % options.prefix
out_gff = "%s.gff" % options.prefix
if os.path.exists(full_fq):
os.remove(full_fq)
if os.path.exists(clean_fq):
os.remove(clean_fq)
pre = fasta.read_precursor(options.fa, "")
mir = mapper.read_gtf_to_precursor(options.gtf)
nt = ['A', 'T', 'G', 'C']
gffs = dict()
h = header.create(["sampleX"], "miRBase1", "")
for precursor in pre:
seq = pre[precursor]
gffs.update(create_iso(precursor, mir, seq, options.numsim, options.exp))
_write(gffs, h, out_gff)