def test04(self):
with BZ2File(self.data('rle_100_left.ssearch.bz2')) as f:
aligns = list(sequtils.parse_ssearch36(f))
# 100 total sequences
self.assertEqual(len(set(a['q_name'] for a in aligns)), 100)
# searched against 4 primers
self.assertEqual(len(aligns), 400)
def action(args):
extras = parse_extras(args.extra_fields) if args.extra_fields else {}
aligns = islice(parse_ssearch36(args.alignments, False), args.limit)
if args.min_zscore:
aligns = (a for a in aligns
if float(a['sw_zscore']) >= args.min_zscore)
aligns = groupby(aligns, key=itemgetter('q_name'))
if args.top_alignment:
aligns = (next(a) for _, a in aligns)
else:
aligns = (a for _, i in aligns for a in i) # flatten groupby iters
if args.rlefile:
decoding = {k: v for d in args.rlefile for k, v in d.items()}
def decode(aligns):
aligns['t_seq'], aligns['q_seq'] = homodecodealignment(
aligns['t_seq'], from_ascii(decoding[aligns['t_name']]),
aligns['q_seq'], from_ascii(decoding[aligns['q_name']]))
return aligns
aligns = imap(decode, aligns)
if args.print_one:
pprint.pprint(aligns.next())
sys.exit()
if args.with_diff:
aligns = imap(add_diff, aligns)
if args.fieldnames:
fieldnames = args.fieldnames
else:
# peek at first row fieldnames
top = next(aligns, {})
fieldnames = top.keys()
aligns = chain([top], aligns)
if extras:
fieldnames += extras.keys()
aligns = (dict(d, **extras) for d in aligns)
writer = csv.DictWriter(args.out,
extrasaction='ignore',
fieldnames=fieldnames)
if args.header:
writer.writeheader()
for a in aligns:
writer.writerow(a)
def test03(self):
"""
No output name, clean up
"""
q, t = self.data('two.fasta'), self.data('ten.fasta')
with sequtils.run_ssearch(q, t) as aligns:
parsed = sequtils.parse_ssearch36(aligns)
self.assertEqual(set(['H59735', 'T70875']),
{d['q_name'] for d in parsed})
# should not still exist
self.assertFalse(path.exists(aligns.name))
def test02(self):
"""
No output name, don't clean up
"""
q, t = self.data('two.fasta'), self.data('ten.fasta')
with sequtils.run_ssearch(q, t, cleanup=False) as aligns:
parsed = sequtils.parse_ssearch36(aligns)
self.assertEqual(set(['H59735', 'T70875']),
{d['q_name'] for d in parsed})
# should still exist since we specified 'cleanup=False'
self.assertTrue(path.exists(aligns.name))
def test01(self):
"""
Provide an output name
"""
q, t = self.data('two.fasta'), self.data('ten.fasta')
out = path.join(self.mkoutdir(), 'aligns.ssearch')
with sequtils.run_ssearch(q, t, out) as aligns:
self.assertEqual(out, aligns.name)
parsed = sequtils.parse_ssearch36(aligns)
self.assertEqual(set(['H59735', 'T70875']),
{d['q_name'] for d in parsed})
# should still exist since we provided a name for the output
self.assertTrue(path.exists(out))
def action(args):
# setup ssearch command and communicate
command = ["ssearch36"]
command += ["-m", "10"] # 10 is parseable key:value output format
command += ["-n"] # DNA/RNA query
command += ["-z", args.statistical_calculation]
command += ["-g", args.gap_extension_penalty]
command += ["-f", args.gap_open_penalty]
command += ["-T", str(args.threads)]
if args.strand == "forward":
command += ["-3"] # forward strand only
if args.full_sequences:
command += ["-a"]
if not args.all_alignments:
command += ["-b", "1"]
command += ["-d", "1"]
command += [args.query, args.library]
# If query or library file is empty, don't bother executing ssearch.
# Just print empty file
# with a header and exit
if os.stat(args.query).st_size == 0 or os.stat(args.library).st_size == 0:
# write empty header
if args.fieldnames:
fieldnames = args.fieldnames
if fieldnames:
writer = DictWriter(args.out, extrasaction="ignore", fieldnames=fieldnames)
if args.header:
writer.writeheader()
return
log.info(" ".join(command))
ssearch = Popen(command, stdout=PIPE, stderr=PIPE)
# parse alignments
aligns = parse_ssearch36(ssearch.stdout)
aligns = (a for a in aligns if float(a["sw_zscore"]) >= args.min_zscore)
aligns = groupby(aligns, key=itemgetter("q_name"))
aligns = (a for _, i in aligns for a in i) # flatten groupby iters
# decode if appropriate
if args.decode:
decoding = {k: v for d in args.decode for k, v in d.items()}
def decode(aligns):
aligns["t_seq"], aligns["q_seq"] = homodecodealignment(
aligns["t_seq"],
from_ascii(decoding[aligns["t_name"]]),
aligns["q_seq"],
from_ascii(decoding[aligns["q_name"]]),
)
return aligns
aligns = imap(decode, aligns)
# calculate coverage for each item and repack into generator
# coverage = |query alignment| / |query length|
aligns = (
dict(d, qcovs=str((float(d["q_al_stop"]) - float(d["q_al_start"])) / float(d["q_sq_len"]))) for d in aligns
)
# write results
if args.fieldnames:
fieldnames = args.fieldnames
else:
# peek at first row fieldnames
top = next(aligns, {})
fieldnames = top.keys()
if top:
aligns = chain([top], aligns)
writer = DictWriter(args.out, extrasaction="ignore", fieldnames=fieldnames)
if args.header:
writer.writeheader()
for a in aligns:
writer.writerow(a)
error = set(e.strip() for e in ssearch.stderr)
error = ", ".join(error)
if ssearch.wait() != 0:
raise CalledProcessError(ssearch.returncode, error)
if error:
log.error(error)
def action(args):
# setup ssearch command and communicate
command = ['fasta36']
command += ['-m', '10']
command += ['-3']
command += ['-n']
command += ['-g', args.gap_extension_penalty]
command += ['-f', args.gap_open_penalty]
command += ['-T', str(args.threads)]
if args.full_sequences:
command += ['-a']
if not args.all_alignments:
command += ['-b', '1']
command += ['-d', '1']
command += [args.query, args.library]
log.info(' '.join(command))
pipe = Popen(command, stdout = PIPE, stderr = PIPE)
# parse alignments
aligns = parse_ssearch36(pipe.stdout)
aligns = (a for a in aligns if float(a['fa_zscore']) >= args.min_zscore)
aligns = groupby(aligns, key = itemgetter('q_name'))
aligns = (a for _,i in aligns for a in i) # flatten groupby iters
# decode if appropriate
if args.decode:
decoding = {k:v for d in args.decode for k,v in d.items()}
def decode(aligns):
aligns['t_seq'], aligns['q_seq'] = homodecodealignment(
aligns['t_seq'], from_ascii(decoding[aligns['t_name']]),
aligns['q_seq'], from_ascii(decoding[aligns['q_name']]))
return aligns
aligns = imap(decode, aligns)
# write results
if args.fieldnames:
fieldnames = args.fieldnames
else:
# peek at first row fieldnames
top = next(aligns, {})
fieldnames = top.keys()
if top:
aligns = chain([top], aligns)
if fieldnames:
writer = DictWriter(args.out,
extrasaction = 'ignore',
fieldnames = fieldnames)
if args.header:
writer.writeheader()
for a in aligns:
writer.writerow(a)
error = set(e.strip() for e in pipe.stderr)
error = ', '.join(error)
if pipe.wait() != 0:
raise CalledProcessError(pipe.returncode, error)
if error:
log.error(error)
def action(args):
# setup ssearch command and communicate
command = ['fasta36']
command += ['-m', '10']
command += ['-3']
command += ['-n']
command += ['-g', args.gap_extension_penalty]
command += ['-f', args.gap_open_penalty]
command += ['-T', str(args.threads)]
if args.full_sequences:
command += ['-a']
if not args.all_alignments:
command += ['-b', '1']
command += ['-d', '1']
command += [args.query, args.library]
log.info(' '.join(command))
pipe = Popen(command, stdout=PIPE, stderr=PIPE)
# parse alignments
aligns = parse_ssearch36(pipe.stdout)
aligns = (a for a in aligns if float(a['fa_zscore']) >= args.min_zscore)
aligns = groupby(aligns, key=itemgetter('q_name'))
aligns = (a for _, i in aligns for a in i) # flatten groupby iters
# decode if appropriate
if args.decode:
decoding = {k: v for d in args.decode for k, v in d.items()}
def decode(aligns):
aligns['t_seq'], aligns['q_seq'] = homodecodealignment(
aligns['t_seq'], from_ascii(decoding[aligns['t_name']]),
aligns['q_seq'], from_ascii(decoding[aligns['q_name']]))
return aligns
aligns = imap(decode, aligns)
# write results
if args.fieldnames:
fieldnames = args.fieldnames
else:
# peek at first row fieldnames
top = next(aligns, {})
fieldnames = top.keys()
if top:
aligns = chain([top], aligns)
if fieldnames:
writer = DictWriter(args.out,
extrasaction='ignore',
fieldnames=fieldnames)
if args.header:
writer.writeheader()
for a in aligns:
writer.writerow(a)
error = set(e.strip() for e in pipe.stderr)
error = ', '.join(error)
if pipe.wait() != 0:
raise CalledProcessError(pipe.returncode, error)
if error:
log.error(error)
def action(args):
# setup ssearch command and communicate
command = ['ssearch36']
command += ['-m', '10'] # 10 is parseable key:value output format
command += ['-n'] # DNA/RNA query
command += ['-z', args.statistical_calculation]
command += ['-g', args.gap_extension_penalty]
command += ['-f', args.gap_open_penalty]
command += ['-T', str(args.threads)]
if args.strand == 'forward':
command += ['-3'] # forward strand only
if args.full_sequences:
command += ['-a']
if not args.all_alignments:
command += ['-b', '1']
command += ['-d', '1']
command += [args.query, args.library]
# If query or library file is empty, don't bother executing ssearch.
# Just print empty file
# with a header and exit
if os.stat(args.query).st_size == 0 or os.stat(args.library).st_size == 0:
# write empty header
if args.fieldnames:
fieldnames = args.fieldnames
if fieldnames:
writer = DictWriter(args.out,
extrasaction='ignore',
fieldnames=fieldnames)
if args.header:
writer.writeheader()
return
log.info(' '.join(command))
ssearch = Popen(command, stdout=PIPE, stderr=PIPE)
# parse alignments
aligns = parse_ssearch36(ssearch.stdout)
aligns = (a for a in aligns if float(a['sw_zscore']) >= args.min_zscore)
aligns = groupby(aligns, key=itemgetter('q_name'))
aligns = (a for _, i in aligns for a in i) # flatten groupby iters
# decode if appropriate
if args.decode:
decoding = {k: v for d in args.decode for k, v in d.items()}
def decode(aligns):
aligns['t_seq'], aligns['q_seq'] = homodecodealignment(
aligns['t_seq'], from_ascii(decoding[aligns['t_name']]),
aligns['q_seq'], from_ascii(decoding[aligns['q_name']]))
return aligns
aligns = imap(decode, aligns)
# calculate coverage for each item and repack into generator
# coverage = |query alignment| / |query length|
aligns = (dict(d, qcovs=str(
(float(d['q_al_stop']) - float(d['q_al_start'])) /
float(d['q_sq_len'])
)) for d in aligns)
# write results
if args.fieldnames:
fieldnames = args.fieldnames
else:
# peek at first row fieldnames
top = next(aligns, {})
fieldnames = top.keys()
if top:
aligns = chain([top], aligns)
writer = DictWriter(args.out,
extrasaction='ignore',
fieldnames=fieldnames)
if args.header:
writer.writeheader()
for a in aligns:
writer.writerow(a)
error = set(e.strip() for e in ssearch.stderr)
error = ', '.join(error)
if ssearch.wait() != 0:
raise CalledProcessError(ssearch.returncode, error)
if error:
log.error(error)
