def fusions_in_blocks(self, blocks):
# TODO use exons instead of genes
fusions = []
for block in blocks[:-1]:
chrom_id = block['end']['chrom']
end = block['end']['pos']
fusion = {}
fusion['start gene'] = ensembl_requests.get_genes(chrom_id, end, end)
fusion['start is cut'] = fusion['start gene'] and len(fusion['start gene']) > 0;
# Exons never seem to overlap a cut, so leaving this out for now
# exons = ensembl_requests.get_exons(chrom_id, end, end)
# filtered_exons = [e for e in exons if (e['start'] <= end) and (e['end'] >= end)]
# fusion['start cut at exon'] = len(filtered_exons) > 0
fusions.append(fusion)
for i, block in enumerate(blocks[1:]):
chrom_id = block['start']['chrom']
start = block['start']['pos']
fusions[i]['end gene'] = ensembl_requests.get_genes(chrom_id, start, start)
fusions[i]['end is cut'] = fusions[i]['end gene'] and len(fusions[i]['end gene']) > 0;
# Exons never seem to overlap a cut, so leaving this out for now
# exons = ensembl_requests.get_exons(chrom_id, start, start)
# filtered_exons = [e for e in exons if (e['start'] <= start) and (e['end'] >= start)]
# fusions[i]['end cut at exon'] = len(filtered_exons) > 0
fusions[i]['genes are fused'] = fusions[i]['start is cut'] and fusions[i]['end is cut']
return fusions
def fetch_genes(self, chrom_id, start, end, species='human'):
'''Returns JSON list of genes in a given region, or None if request is
bad. Maximum request region at a time is 5Mb. Ensembl accepts both the
naming format used by UCSC and the one used by Meerkat. Note: No gene
listings are available for the scaffolds and patches, which have names
using the prefix GL-.'''
# get rid of this method - duplicated in web_viz
genes = ensembl_requests.get_genes(species, chrom_id, start, end)
#print json.dumps(genes, indent=4, sort_keys=True)
return genes
def genes_in_blocks(self, blocks):
genes_per_block = []
request_limit = 5000000
for block in blocks:
chrom_id = block['start']['chrom']
start = block['start']['pos']
end = block['end']['pos']
print 'genes in blocks', chrom_id, start, end
if (abs(start - end) > request_limit):
# For now, if the request region exceeds the Ensembl API limit, we just return the
# genes near the ends of the sequence block
start_genes = ensembl_requests.get_genes(chrom_id, start, start+request_limit-1)
end_genes = ensembl_requests.get_genes(chrom_id, end-request_limit+1, end)
start_genes_dict = {g['id']:g for g in start_genes}
end_genes_dict = {g['id']:g for g in end_genes}
start_genes_dict.update(end_genes_dict) # start_genes_dict is now the union of the two dicts (TODO confusing)
genes = list(start_genes_dict.values())
else:
genes = ensembl_requests.get_genes(chrom_id, start, end) if start < end \
else ensembl_requests.get_genes(chrom_id, end, start)
genes_per_block.append(genes);
return genes_per_block
def json_genes(chrom_id, start, end, species='human'):
# TODO may want to move request into javascript
genes = ensembl_requests.get_genes(chrom_id, start, end, species)
return json.dumps(genes)
def json_genes(chrom_id, start, end, species='human'):
# TODO may want to move request into javascript
genes = ensembl_requests.get_genes(chrom_id, start, end, species)
return json.dumps(genes)