def action(args):
if args.inplace and args.infile is sys.stdin:
log.error('Error: cannot use the --inplace option with stdin')
return
if args.rename:
raise NotImplementedError
reader = csv.DictReader(args.infile)
fieldnames = reader.fieldnames or []
new_fields = parse_extras(args.add) if args.add else {}
if new_fields:
fieldnames.extend(new_fields.keys())
reader = imap(lambda row: dict(row, **new_fields), reader)
if args.inplace:
outfile = tmp(args.infile.name)
else:
outfile = args.outfile
with opener(outfile, 'w') as fout:
writer = csv.DictWriter(fout, fieldnames, extrasaction='ignore')
writer.writeheader()
writer.writerows(reader)
if args.inplace:
os.rename(fout.name, args.infile.name)
def build_parser(parser):
parser.add_argument('aligns',
nargs = '?',
default = sys.stdin,
type = Opener('r'),
help = 'csvfile of ssearch results')
parser.add_argument('-o', '--out',
default = sys.stdout,
type = Opener('w'),
help = 'csv file tallying each error category for each read')
parser.add_argument('-m', '--homopolymer-matrix',
dest = 'matrix',
type = lambda f: writer(Opener('w')(f)),
help = 'csv file containing transition matrix of homopolymer lengths')
parser.add_argument('-M', '--homopolymer-max',
default = 6,
type = int,
help = 'csv homopolymer length above which counts are binned')
parser.add_argument('--step',
action = 'store_true',
help = 'step through reults (for debugging)')
parser.add_argument('-f', '--extra-fields',
type = lambda f: parse_extras(f),
default = {},
help="extra fields for csv file in form 'name1:val1,name2:val2'")
parser.add_argument('--output-alignment',
action='store_true',
help = 'Include the actual alignment in csv output')
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 build_parser(parser):
parser.add_argument('aligns',
nargs='?',
default=sys.stdin,
type=Opener('r'),
help='csvfile of ssearch results')
parser.add_argument(
'-o',
'--out',
default=sys.stdout,
type=Opener('w'),
help='csv file tallying each error category for each read')
parser.add_argument(
'-m',
'--homopolymer-matrix',
dest='matrix',
type=lambda f: writer(Opener('w')(f)),
help='csv file containing transition matrix of homopolymer lengths')
parser.add_argument(
'-M',
'--homopolymer-max',
default=6,
type=int,
help='csv homopolymer length above which counts are binned')
parser.add_argument('--step',
action='store_true',
help='step through reults (for debugging)')
parser.add_argument(
'-f',
'--extra-fields',
type=lambda f: parse_extras(f),
default={},
help="extra fields for csv file in form 'name1:val1,name2:val2'")
parser.add_argument('--output-alignment',
action='store_true',
help='Include the actual alignment in csv output')
default=True,
dest='show_header')
BASES = set(['A', 'C', 'G', 'T'])
def action(args):
try:
from numpy import mean
except ImportError, e:
print(e)
sys.exit(1)
extras = parse_extras(args.extra_fields) if args.extra_fields else {}
# TODO: add 'ambiguities'
fieldnames = ['name', 'length', 'mean', 'ambig']
fieldnames += extras.keys()
stats = DictWriter(args.out, fieldnames=fieldnames)
if args.show_header:
stats.writeheader()
for s in islice(SeqIO.parse(args.fastq, 'fastq'), args.limit):
qual = s.letter_annotations["phred_quality"]
ambig = len([1 for b in s.seq if b not in BASES])
row = [s.name.replace(':', '_'), len(s), mean(qual), ambig]
row += extras.values()
