diff galaxy-tools/biobank/tools/check_update_parents_data.py @ 3:43be74e62bfe draft

Uploaded

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/galaxy-tools/biobank/tools/check_update_parents_data.py	Thu Sep 22 08:57:04 2016 -0400
@@ -0,0 +1,113 @@
+import sys, csv, argparse, logging, os
+from collections import Counter
+
+from bl.vl.kb import KnowledgeBase as KB
+import bl.vl.utils.ome_utils as vlu
+from bl.vl.utils import get_logger, LOG_LEVELS
+
+
+def make_parser():
+    parser = argparse.ArgumentParser(description='check data that will be passed to the update_parents tool')
+    parser.add_argument('--logfile', type=str, help='log file (default=stderr)')
+    parser.add_argument('--loglevel', type=str, choices=LOG_LEVELS,
+                        help='logging level (default=INFO)', default='INFO')
+    parser.add_argument('-H', '--host', type=str, help='omero hostname')
+    parser.add_argument('-U', '--user', type=str, help='omero user')
+    parser.add_argument('-P', '--passwd', type=str, help='omero password')
+    parser.add_argument('--in_file', type=str, required=True,
+                        help='input file')
+    parser.add_argument('--out_file', type=str, required=True,
+                        help='output file')
+    return parser
+
+
+def check_row(row, individuals_map, kb, logger):
+    logger.debug('Checking record %r' % row)
+    try:
+        ind = individuals_map[row['individual']]
+        logger.info('%s is a valid Individual ID' % ind.id)
+        if row['father'] != 'None':
+            father = individuals_map[row['father']]
+            logger.info('%s is a valid Individual ID' % father.id)
+            check_gender(father, kb.Gender.MALE)
+            logger.info('Gender check passed')
+        else:
+            logger.info('None value, no check required')
+        if row['mother'] != 'None':
+            mother = individuals_map[row['mother']]
+            logger.info('%s is a valid Individual ID' % mother.id)
+            check_gender(mother, kb.Gender.FEMALE)
+            logger.info('Gender check passed')
+        else:
+            logger.info('None value, no check required')
+        return True
+    except KeyError, ke:
+        logger.error('%s is not a valid Individual ID, rejecting row' % ke)
+        return False
+    except ValueError, ve:
+        logger.error(ve)
+        return False
+
+
+def check_gender(individual, gender):
+    if individual.gender.enum_label() != gender.enum_label():
+        raise ValueError('Gender for individual %s is %s, expected %s, rejecting row' % (individual.id,
+                                                                                         individual.gender.enum_label(),
+                                                                                         gender.enum_label()))
+    else:
+        pass
+
+
+def main(argv):
+    parser = make_parser()
+    args = parser.parse_args(argv)
+
+    logger = get_logger('check_update_parents_data', level=args.loglevel,
+                        filename=args.logfile)
+
+    try:
+        host = args.host or vlu.ome_host()
+        user = args.user or vlu.ome_user()
+        passwd = args.passwd or vlu.ome_passwd()
+    except ValueError, ve:
+        logger.critical(ve)
+        sys.exit(ve)
+
+    kb = KB(driver='omero')(host, user, passwd)
+
+    logger.info('Preloading all individuals from the system')
+    inds = kb.get_objects(kb.Individual)
+    logger.info('%d individuals loaded' % len(inds))
+    inds_lookup = {}
+    for i in inds:
+        inds_lookup[i.id] = i
+
+    with open(args.in_file) as infile, open(args.out_file, 'w') as outfile:
+        reader = csv.DictReader(infile, delimiter='\t')
+        records = list(reader)
+        logger.info('Check for duplicated in \'individual\' column')
+        recs_by_ind = {}
+        for rec in records:
+            recs_by_ind.setdefault(rec['individual'], []).append(rec)
+        ct = Counter()
+        duplicated = []
+        for k,v in recs_by_ind.iteritems():
+            if len(v) > 1:
+                duplicated.append(k)
+        for dupl in duplicated:
+            logger.info('Individual %s is a duplicated' % dupl)
+            for r in recs_by_ind.pop(dupl):
+                logger.info('Removing record %r' % r)
+        good_records = sum(recs_by_ind.itervalues(), [])
+        logger.info('Duplicated check completed')
+        writer = csv.DictWriter(outfile, reader.fieldnames, delimiter='\t')
+        writer.writeheader()
+        logger.info('Checking records')
+        for row in good_records:
+            if check_row(row, inds_lookup, kb, logger):
+                writer.writerow(row)
+                logger.debug('Record %r written in output file' % row)
+        logger.info('Records check completed')
+
+if __name__ == '__main__':
+    main(sys.argv[1:])