Mercurial > repos > galaxyp > percolator

comparison metafiles2pin.py @ 0:3a49065a05d6 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/percolator commit b4871f9659a924a68430aed3a93f4f9bad733fd6
author galaxyp
date Wed, 07 Dec 2016 16:43:51 -0500
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:3a49065a05d6
1 import argparse
2 import os
3 import re
4 from collections import OrderedDict
5
6
7 def get_filename_index_with_identifier(spectrafiles, pool_id):
8 pool_indices = []
9 for index, fn in enumerate(spectrafiles):
10 if re.search(pool_id, fn) is not None:
11 pool_indices.append(index)
12 return pool_indices
13
14
15 def get_perco_batches_from_spectrafiles(spectrafiles, batchsize, ppool_ids):
16 """For an amount of input spectra files, pool identifiers and a batch size,
17 return batches of files that can be percolated together"""
18 if ppool_ids:
19 filegroups = OrderedDict([(p_id, get_filename_index_with_identifier(
20 spectrafiles, p_id)) for p_id in ppool_ids])
21 else:
22 filegroups = {1: range(len(spectrafiles))}
23 batch = []
24 for grouped_indices in filegroups.values():
25 for index in grouped_indices:
26 batch.append(index)
27 if len(batch) == int(batchsize):
28 yield batch
29 batch = []
30 if len(batch) > 0:
31 yield batch
32 batch = []
33
34
35 def main():
36 parser = argparse.ArgumentParser()
37 parser.add_argument('--batchsize', dest='batchsize')
38 parser.add_argument('--spectrafiles', dest='spectrafiles', nargs='+')
39 parser.add_argument('--searchfiles', dest='searchfiles', nargs='+')
40 parser.add_argument('--percolator-pool-ids', dest='percopoolids', nargs='+', default=False)
41 args = parser.parse_args()
42 outpath = os.path.join(os.getcwd(), 'metafiles')
43 os.makedirs(outpath)
44 for count, batch in enumerate(get_perco_batches_from_spectrafiles(
45 args.spectrafiles, args.batchsize, args.percopoolids)):
46 batchfiles = [args.searchfiles[index] for index in batch]
47 out_file = os.path.join(outpath, 'percolatorpool{}.meta2pin'.format(
48 str(count)))
49 with open(out_file, 'w') as fp:
50 fp.write('\n'.join(batchfiles))
51
52
53 if __name__ == '__main__':
54 main()