Mercurial > repos > iuc > gemini_fusions

comparison test-data/util/shrink_simple_tab.py @ 4:b35fd707c586 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/gemini commit 283362494058ed64143b1f27afb447b8a1cb4313
author iuc
date Fri, 14 Dec 2018 12:46:37 -0500
parents
children
comparison
equal deleted inserted replaced
3:91d53f89ca54 4:b35fd707c586
1 from __future__ import print_function
2
3 import argparse
4 from functools import partial
5
6
7 def keep_line(line, pos_cols, region):
8 fields = line.rstrip().split(b'\t')
9 if fields[pos_cols[0]] == region[0]: # same chromosome
10 if (
11 region[1] < int(fields[pos_cols[1]]) < region[2]
12 ) or (
13 region[1] < int(fields[pos_cols[2]]) < region[2]
14 ):
15 return True
16
17
18 def main(infile, ofile, num_header_lines):
19 print(infile, '->', ofile)
20 with open(infile, 'rb') as i:
21 with open(ofile, 'wb') as o:
22 # copy header lines
23 for c in range(num_header_lines):
24 o.write(next(i))
25 for line in i:
26 if keep_line(line):
27 o.write(line)
28
29
30 if __name__ == '__main__':
31 p = argparse.ArgumentParser()
32 p.add_argument('infile')
33 p.add_argument(
34 '-r', '--region',
35 required=True,
36 help='the region of the input file to rewrite'
37 )
38 p.add_argument(
39 '-o', '--ofile',
40 required=True,
41 help="the name of the output file"
42 )
43 p.add_argument(
44 '-c', '--cols',
45 nargs=3, type=int, required=True,
46 help="the columns of the input file specifying chrom, start and stop, "
47 "respectively"
48 )
49 p.add_argument(
50 '-n', '--num-header-lines',
51 type=int, default=0,
52 help='the number of header lines present in the input; These will '
53 'always be copied over to the new file.'
54 )
55 args = vars(p.parse_args())
56
57 chrom, reg = args['region'].split(':')
58 region = [chrom.encode()] + [int(x) for x in reg.split('-')]
59 keep_line = partial(keep_line, pos_cols=args['cols'], region=region)
60
61 main(args['infile'], args['ofile'], args['num_header_lines'])