def yield_ensembl_names(gene_ids):
import biopsy.identifiers.biomart as biomart, csv
query = biomart.new_query()
dataset = biomart.add_dataset(query, 'mmusculus_gene_ensembl')
biomart.add_filter(dataset, 'ensembl_gene_id', ",".join(imap(str, gene_ids)))
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'external_gene_id')
for row in biomart.yield_csv_query_results(query):
yield row[0], row[1]
def yield_ensembl_names(gene_ids):
import biopsy.identifiers.biomart as biomart, csv
query = biomart.new_query()
dataset = biomart.add_dataset(query, 'mmusculus_gene_ensembl')
biomart.add_filter(dataset, 'ensembl_gene_id',
",".join(imap(str, gene_ids)))
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'external_gene_id')
for row in biomart.yield_csv_query_results(query):
yield row[0], row[1]
def yield_uniprot_gene_mappings(gene_ids):
import biopsy.identifiers.biomart as biomart, csv
query = biomart.new_query()
dataset = biomart.add_dataset(query, 'mmusculus_gene_ensembl')
biomart.add_filter(dataset, 'ensembl_gene_id', ",".join(gene_ids))
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'unified_uniprot_accession')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
if row[1]:
yield biopsy.DbRef.parse_as(row[0], biopsy.db.ensembl), biopsy.DbRef.parse_as(row[1], biopsy.db.swissprot)
def proteins_for_species(species):
"Goes to biomart to map all genes in species to Swissprot proteins"
from . import entrez
result = cookbook.DictOfSets()
# build biomart query for swissprot and execute
query = biomart.new_query()
dataset = biomart.add_dataset(query, species)
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'ensembl_transcript_id')
biomart.add_attribute(dataset, 'uniprot_swissprot_accession')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
gene_ref = biopsy.transfac.DbRef.parse_as(row[0],
biopsy.transfac.db.ensembl)
transcript_ref = biopsy.transfac.DbRef.parse_as(
row[1], biopsy.transfac.db.ensembl)
if row[2]:
protein_ref = biopsy.transfac.DbRef.parse_as(
row[2], biopsy.transfac.db.swissprot)
result[gene_ref].add((transcript_ref, protein_ref))
# build biomart query for entrez protein and execute
query = biomart.new_query()
dataset = biomart.add_dataset(query, species)
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'ensembl_transcript_id')
biomart.add_attribute(dataset, 'protein')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
gene_ref = biopsy.transfac.DbRef.parse_as(row[0],
biopsy.transfac.db.ensembl)
transcript_ref = biopsy.transfac.DbRef.parse_as(
row[1], biopsy.transfac.db.ensembl)
if row[2]:
protein_acc = row[2]
#print "'%s'" % protein_acc
if protein_acc in entrez.mouse_proteins().acc2id:
for protein_id in entrez.mouse_proteins().acc2id[protein_acc]:
ref = T.DbRef(T.db.entrez_protein, "", protein_id)
result[gene_ref].add((transcript_ref, ref))
return result
def yield_mouse_orthologs(hs_genes):
# map into mouse orthologs using biomart
query = B.new_query()
dataset = B.add_dataset(query, 'hsapiens_gene_ensembl')
B.add_attribute(dataset, 'ensembl_gene_id')
B.add_attribute(dataset, 'mouse_ensembl_gene')
filter = B.add_filter(dataset, name='ensembl_gene_id', value='')
filter.set('value', ','.join(ensembl_hs_genes))
for chunk in B.split_big_list(ensembl_hs_genes, 50):
#logging.info('Querying Ensembl biomart for chunk of %d genes', len(chunk))
filter.set('value', ','.join(chunk))
for row in B.yield_csv_query_results(query):
if row[1]:
yield row[1]
def yield_mouse_orthologs(hs_genes):
# map into mouse orthologs using biomart
query = B.new_query()
dataset = B.add_dataset(query, 'hsapiens_gene_ensembl')
B.add_attribute(dataset, 'ensembl_gene_id')
B.add_attribute(dataset, 'mouse_ensembl_gene')
filter = B.add_filter(dataset, name='ensembl_gene_id', value='')
filter.set('value', ','.join(ensembl_hs_genes))
for chunk in B.split_big_list(ensembl_hs_genes, 50):
#logging.info('Querying Ensembl biomart for chunk of %d genes', len(chunk))
filter.set('value', ','.join(chunk))
for row in B.yield_csv_query_results(query):
if row[1]:
yield row[1]
def proteins_for_species(species):
"Goes to biomart to map all genes in species to Swissprot proteins"
from . import entrez
result = cookbook.DictOfSets()
# build biomart query for swissprot and execute
query = biomart.new_query()
dataset = biomart.add_dataset(query, species)
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'ensembl_transcript_id')
biomart.add_attribute(dataset, 'uniprot_swissprot_accession')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
gene_ref = biopsy.transfac.DbRef.parse_as(row[0], biopsy.transfac.db.ensembl)
transcript_ref = biopsy.transfac.DbRef.parse_as(row[1], biopsy.transfac.db.ensembl)
if row[2]:
protein_ref = biopsy.transfac.DbRef.parse_as(row[2], biopsy.transfac.db.swissprot)
result[gene_ref].add((transcript_ref, protein_ref))
# build biomart query for entrez protein and execute
query = biomart.new_query()
dataset = biomart.add_dataset(query, species)
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'ensembl_transcript_id')
biomart.add_attribute(dataset, 'protein')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
gene_ref = biopsy.transfac.DbRef.parse_as(row[0], biopsy.transfac.db.ensembl)
transcript_ref = biopsy.transfac.DbRef.parse_as(row[1], biopsy.transfac.db.ensembl)
if row[2]:
protein_acc = row[2]
#print "'%s'" % protein_acc
if protein_acc in entrez.mouse_proteins().acc2id:
for protein_id in entrez.mouse_proteins().acc2id[protein_acc]:
ref = T.DbRef(T.db.entrez_protein, "", protein_id)
result[gene_ref].add((transcript_ref, ref))
return result
def _mgi_ids_from_biomart():
"Goes to biomart to map all mouse genes to MGI ids"
from . import entrez
result = cookbook.DictOfSets()
# build biomart query for swissprot and execute
query = biomart.new_query()
dataset = biomart.add_dataset(query, 'mmusculus_gene_ensembl')
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'external_gene_id')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
gene_ref = biopsy.transfac.DbRef.parse_as(row[0], biopsy.transfac.db.ensembl)
if row[1]:
mgi_acc = row[1]
if mgi_acc in mgi.acc2id():
yield gene_ref, mgi.acc2id()[mgi_acc]
def get_orthologs(species1, species2):
"Returns a dict mapping ensembl genes of species 1 to genes of species 2."
result = cookbook.DictOfSets()
# build biomart query and execute
query = biomart.new_query()
dataset = biomart.add_dataset(query, species[species2].dataset())
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, '%s_ensembl_gene' % species[species1].short_name)
biomart.print_query(query, 'query.xml')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
if row[1]:
ref_1 = biopsy.transfac.DbRef.parse_as(row[0], biopsy.transfac.db.ensembl)
ref_2 = biopsy.transfac.DbRef.parse_as(row[1], biopsy.transfac.db.ensembl)
result[ref_2].add(ref_1)
return result
def _mgi_ids_from_biomart():
"Goes to biomart to map all mouse genes to MGI ids"
from . import entrez
result = cookbook.DictOfSets()
# build biomart query for swissprot and execute
query = biomart.new_query()
dataset = biomart.add_dataset(query, 'mmusculus_gene_ensembl')
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, 'external_gene_id')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
gene_ref = biopsy.transfac.DbRef.parse_as(row[0],
biopsy.transfac.db.ensembl)
if row[1]:
mgi_acc = row[1]
if mgi_acc in mgi.acc2id():
yield gene_ref, mgi.acc2id()[mgi_acc]
def get_orthologs(species1, species2):
"Returns a dict mapping ensembl genes of species 1 to genes of species 2."
result = cookbook.DictOfSets()
# build biomart query and execute
query = biomart.new_query()
dataset = biomart.add_dataset(query, species[species2].dataset())
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset,
'%s_ensembl_gene' % species[species1].short_name)
biomart.print_query(query, 'query.xml')
for row in csv.reader(biomart.execute_query(query), delimiter=','):
if row[1]:
ref_1 = biopsy.transfac.DbRef.parse_as(row[0],
biopsy.transfac.db.ensembl)
ref_2 = biopsy.transfac.DbRef.parse_as(row[1],
biopsy.transfac.db.ensembl)
result[ref_2].add(ref_1)
return result
def get_ensembl_go_annotations(genes):
"@return: A map from the given genes to sets of go annotations."
import biopsy.identifiers.biomart as biomart
logging.info('Querying Ensembl biomart for GO annotations of %d genes', len(genes))
result = cookbook.DictOfLists()
for id_attr, evidence_attr in [
('go_biological_process_id', 'go_biological_process_linkage_type'),
('go_cellular_component_id', 'go_cellular_component_linkage_type'),
('go_molecular_function_id', 'go_molecular_function_linkage_type'),
]:
query = biomart.new_query()
dataset = biomart.add_dataset(query, 'mmusculus_gene_ensembl')
biomart.add_attribute(dataset, 'ensembl_gene_id')
biomart.add_attribute(dataset, id_attr)
biomart.add_attribute(dataset, evidence_attr)
filter = biomart.add_filter(dataset, name='ensembl_gene_id', value='')
for chunk in biomart.split_big_list((str(g) for g in genes), 50):
#logging.info('Querying Ensembl biomart for chunk of %d genes', len(chunk))
filter.set('value', ','.join(chunk))
for row in biomart.yield_csv_query_results(query):
if row[2] not in options.go_evidence_codes_to_ignore:
result[row[0]].append(row[1])
logging.info('Found %d go annotations', sum(len(v) for v in result.values()))
return result