def _fn(x):
out = {}
if x[0].strip() not in ['', '\\N']:
out['symbol'] = x[0].strip()
if x[1].strip() not in ['', '\\N']:
_name = SubStr(x[1].strip(), '', ' [Source:').strip()
if _name:
out['name'] = _name
return out
def _fn(x):
import logging
out = {'taxid': int(x[0])}
if x[1].strip() not in ['', '\\N']:
out['symbol'] = x[1].strip()
if x[2].strip() not in ['', '\\N']:
_name = SubStr(x[2].strip(), '', ' [Source:').strip()
if _name:
out['name'] = _name
return out
def get_geneid(self, rec):
'''Return geneid as integer, None if not found.'''
geneid = None
gene_feature = [x for x in rec.features if x.type == 'gene']
# NCBI has now fixed this issue (https://twitter.com/kdpru/status/474673626730741761)
# if len(gene_feature) == 0 and rec.id == 'NR_001526.1':
# print "Known error for NR_001526.1. Fixed."
# return '252949' # a temp fix for this wrong rec from NCBI
assert len(gene_feature) == 1, '#: {}, id: {}'.format(len(gene_feature), rec.id)
gene_feature = gene_feature[0]
db_xref = gene_feature.qualifiers.get('db_xref', None)
if db_xref:
x = [x for x in db_xref if x.startswith('GeneID:')]
if len(x) == 1:
geneid = int(SubStr(x[0], 'GeneID:'))
return geneid
def get_summary(self, rec):
'''Return summary string if available, return '' otherwise.'''
summary = ''
comment = rec.annotations.get('comment', None)
if comment:
if comment.find('Summary:') != -1:
summary = SubStr(comment, 'Summary: ',).replace('\n', ' ')
for end_str in [# '[provided by RefSeq].',
#'[provided by ',
#'[supplied by ',
'##',
# '[RGD',
'COMPLETENESS:',
'Sequence Note:',
'Transcript Variant:',
'CCDS Note:',
'Publication Note:',
' '*10]:
if summary.find(end_str) != -1:
summary = SubStr(summary, end_string=end_str)
summary = summary.strip()
return summary