def _load_genes(self):
self._db.drop_collection('genes')
self._db.genes.ensure_index('gene_id')
self._db.drop_collection('transcripts')
self._db.transcripts.ensure_index('transcript_id')
self._db.transcripts.ensure_index('gene_id')
self._db.drop_collection('exons')
self._db.exons.ensure_index('exon_id')
self._db.exons.ensure_index('gene_id')
gencode_file = gzip.open(self.settings_module.gencode_gtf_file)
size = os.path.getsize(self.settings_module.gencode_gtf_file)
progress = get_progressbar(size, 'Loading gene definitions from GTF')
for datatype, obj in get_data_from_gencode_gtf(gencode_file):
progress.update(gencode_file.fileobj.tell())
if datatype == 'gene':
gene_id = obj['gene_id']
obj['symbol'] = obj['gene_name']
obj['tags'] = {}
# TODO
#obj['coding_size'] = loading_utils.get_coding_size_from_gene_structure(obj)
obj['coding_size'] = 0
self._db.genes.insert(obj)
if datatype == 'transcript':
transcript_id = obj['transcript_id']
obj['tags'] = {}
self._db.transcripts.insert(obj)
if datatype == 'exon':
exon_id = obj['exon_id']
transcript_id = obj['transcript_id']
del obj['transcript_id']
if self._db.exons.find_one({'exon_id': exon_id}):
self._db.exons.update({'exon_id': exon_id}, {'$push': {'transcripts': transcript_id}})
else:
obj['transcripts'] = [transcript_id,]
obj['tags'] = {}
self._db.exons.insert(obj)
if datatype == 'cds':
exon_id = obj['exon_id']
# this works because cds always comes after exon
# this is obviously an inglorious hack - all the gtf parsing should be improved
self._db.exons.update({'exon_id': exon_id}, {'$set': {
'cds_start': obj['start'],
'cds_stop': obj['stop'],
'cds_xstart': obj['xstart'],
'cds_xstop': obj['xstop'],
}})
def _load_genes(self):
self._db.drop_collection('genes')
self._db.genes.ensure_index('gene_id')
self._db.drop_collection('transcripts')
self._db.transcripts.ensure_index('transcript_id')
self._db.transcripts.ensure_index('gene_id')
self._db.drop_collection('exons')
self._db.exons.ensure_index('exon_id')
self._db.exons.ensure_index('gene_id')
gencode_file = gzip.open(self.settings_module.gencode_gtf_file)
size = os.path.getsize(self.settings_module.gencode_gtf_file)
progress = get_progressbar(size, 'Loading gene definitions from GTF')
for datatype, obj in get_data_from_gencode_gtf(gencode_file):
progress.update(gencode_file.fileobj.tell())
if datatype == 'gene':
gene_id = obj['gene_id']
obj['symbol'] = obj['gene_name']
obj['tags'] = {}
# TODO
#obj['coding_size'] = loading_utils.get_coding_size_from_gene_structure(obj)
obj['coding_size'] = 0
self._db.genes.insert(obj)
if datatype == 'transcript':
transcript_id = obj['transcript_id']
obj['tags'] = {}
self._db.transcripts.insert(obj)
if datatype == 'exon':
exon_id = obj['exon_id']
transcript_id = obj['transcript_id']
del obj['transcript_id']
if self._db.exons.find_one({'exon_id': exon_id}):
self._db.exons.update({'exon_id': exon_id}, {'$push': {'transcripts': transcript_id}})
else:
obj['transcripts'] = [transcript_id,]
obj['tags'] = {}
self._db.exons.insert(obj)
if datatype == 'cds':
exon_id = obj['exon_id']
# this works because cds always comes after exon
# this is obviously an inglorious hack - all the gtf parsing should be improved
self._db.exons.update({'exon_id': exon_id}, {'$set': {
'cds_start': obj['start'],
'cds_stop': obj['stop'],
'cds_xstart': obj['xstart'],
'cds_xstop': obj['xstop'],
}})