Mercurial > repos > george-weingart > lefse

comparison home/ubuntu/lefse_to_export/lefse2circlader.py @ 1:db64b6287cd6 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Modified datatypes
author george-weingart
date Wed, 20 Aug 2014 16:56:51 -0400
parents
children
comparison
equal deleted inserted replaced
0:e7cd19afda2e 1:db64b6287cd6
1 #!/usr/bin/env python
2
3 from __future__ import with_statement
4
5 import sys
6 import os
7 import argparse
8
9 def read_params(args):
10 parser = argparse.ArgumentParser(description='Convert LEfSe output to '
11 'Circlader input')
12 parser.add_argument( 'inp_f', metavar='INPUT_FILE', nargs='?',
13 default=None, type=str,
14 help="the input file [stdin if not present]")
15 parser.add_argument( 'out_f', metavar='OUTPUT_FILE', nargs='?',
16 default=None, type=str,
17 help="the output file [stdout if not present]")
18 parser.add_argument('-l', metavar='levels with label', default=0, type=int)
19
20 return vars(parser.parse_args())
21
22 def lefse2circlader(par):
23 finp,fout = bool(par['inp_f']), bool(par['out_f'])
24
25 with open(par['inp_f']) if finp else sys.stdin as inpf:
26 put_bm = (l.strip().split('\t') for l in inpf.readlines())
27 biomarkers = [p for p in put_bm if len(p) > 2]
28
29 circ = [ [ b[0],
30 "" if b[0].count('.') > par['l'] else b[0].split('.')[-1],
31 b[2],
32 b[2]+"_col" ] for b in biomarkers]
33
34 with open(par['out_f'],'w') if fout else sys.stdout as out_file:
35 for c in circ:
36 out_file.write( "\t".join( c ) + "\n" )
37
38 if __name__ == '__main__':
39 params = read_params(sys.argv)
40 lefse2circlader(params)
41
42
43