|
3
|
1 # This tool format output files from kb_query vessels_by_individual
|
|
|
2 # into a tabular format with all data related to an individual grouped
|
|
|
3 # in each row. The tool needs as input a mapping file like
|
|
|
4 #
|
|
|
5 # individual_id label
|
|
|
6 # V12311 A_STUDY:A_CODE
|
|
|
7 # V135115 A_STUDY:B_CODE
|
|
|
8 #
|
|
|
9 # in order to use a known label and not VIDs for each row
|
|
|
10
|
|
|
11 import csv, sys, argparse, logging
|
|
|
12
|
|
|
13 LOG_FORMAT = '%(asctime)s|%(levelname)-8s|%(message)s'
|
|
|
14 LOG_DATEFMT = '%Y-%m-%d %H:%M:%S'
|
|
|
15 LOG_LEVELS = ['DEBUG', 'INFO', 'WARNING', 'ERROR', 'CRITICAL']
|
|
|
16
|
|
|
17 def make_parser():
|
|
|
18 parser = argparse.ArgumentParser(description='format kb_query vessels_by_individual output file to tabular format')
|
|
|
19 parser.add_argument('--logfile', type=str, help='log file (default=stderr)')
|
|
|
20 parser.add_argument('--loglevel', type=str, choices=LOG_LEVELS,
|
|
|
21 help='logging level', default='INFO')
|
|
|
22 parser.add_argument('--in_file', type=str, required=True,
|
|
|
23 help='input file (obtained using kb_query vessels by individual tool)')
|
|
|
24 parser.add_argument('--map_file', type=str, required=True,
|
|
|
25 help='mapping file')
|
|
|
26 parser.add_argument('--out_file', type=str, required=True,
|
|
|
27 help='output file')
|
|
|
28 return parser
|
|
|
29
|
|
|
30 def get_mapping(records, grouper_field, grouped_field):
|
|
|
31 mapping = {}
|
|
|
32 for rec in records:
|
|
|
33 mapping.setdefault(rec[grouper_field], []).append(rec[grouped_field])
|
|
|
34 return mapping
|
|
|
35
|
|
|
36 def get_labels_mapping(reader, logger):
|
|
|
37 rows = [r for r in reader]
|
|
|
38 lmap = get_mapping(rows, 'individual', 'label')
|
|
|
39 logger.info('%d labels grouped for %d individuals' % (len(rows),
|
|
|
40 len(lmap)))
|
|
|
41 return lmap
|
|
|
42
|
|
|
43 def get_vessels_mapping(reader, logger):
|
|
|
44 rows = [r for r in reader]
|
|
|
45 vmap = get_mapping(rows, 'individual', 'vessel_label')
|
|
|
46 logger.info('%d vessels grouped for %d individuals' % (len(rows),
|
|
|
47 len(vmap)))
|
|
|
48 return vmap
|
|
|
49
|
|
|
50 def build_record(label, vessels):
|
|
|
51 record = {'individual_label' : '--'.join(label)}
|
|
|
52 for v in vessels:
|
|
|
53 record['vessel_%d' % (vessels.index(v) + 1)] = v
|
|
|
54 return record
|
|
|
55
|
|
|
56 def main(argv):
|
|
|
57 parser = make_parser()
|
|
|
58 args = parser.parse_args(argv)
|
|
|
59
|
|
|
60 log_level = getattr(logging, args.loglevel)
|
|
|
61 kwargs = {'format' : LOG_FORMAT,
|
|
|
62 'datefmt' : LOG_DATEFMT,
|
|
|
63 'level' : log_level}
|
|
|
64 if args.logfile:
|
|
|
65 kwargs['filename'] = args.logfile
|
|
|
66 logging.basicConfig(**kwargs)
|
|
|
67 logger = logging.getLogger()
|
|
|
68
|
|
|
69 with open(args.map_file) as mf:
|
|
|
70 reader = csv.DictReader(mf, delimiter='\t')
|
|
|
71 labels_map = get_labels_mapping(reader, logger)
|
|
|
72
|
|
|
73 with open(args.in_file) as inf:
|
|
|
74 reader = csv.DictReader(inf, delimiter='\t')
|
|
|
75 vessels_map = get_vessels_mapping(reader, logger)
|
|
|
76
|
|
|
77 max_vessels_count = max([len(v) for v in vessels_map.values()])
|
|
|
78 csv_fields = ['individual_label']
|
|
|
79 for x in xrange(max_vessels_count):
|
|
|
80 csv_fields.append('vessel_%d' % (x+1))
|
|
|
81
|
|
|
82 with open(args.out_file, 'w') as ofile:
|
|
|
83 writer = csv.DictWriter(ofile, csv_fields, delimiter='\t')
|
|
|
84 writer.writeheader()
|
|
|
85 for ind, vessels in vessels_map.iteritems():
|
|
|
86 writer.writerow(build_record(labels_map[ind], vessels))
|
|
|
87
|
|
|
88 logger.info('Job completed')
|
|
|
89
|
|
|
90 if __name__ == '__main__':
|
|
|
91 main(sys.argv[1:])
|
