def buildIdentifierMap(query_species):
columns = ('ensembl_gene_id',
'entrezgene',
'hgnc_id',
'hgnc_symbol')
data = Biomart.biomart_iterator(
columns,
dataset="%s_gene_ensembl" % query_species)
map_identifiers = collections.defaultdict(set)
for row in data:
ensid = row['ensembl_gene_id']
for column in columns[1:]:
xid = str(row[column])
if xid == "NA" or xid == "":
continue
if ensid.startswith('LRG_'):
continue
map_identifiers[xid].add((ensid, column))
# convert to lists
map_identifiers = dict([(x, list(y)) for x, y in list(map_identifiers.items())])
return map_identifiers
def buildIdentifierMap(query_species):
columns = ('ensembl_gene_id',
'entrezgene',
'hgnc_id',
'hgnc_symbol')
data = Biomart.biomart_iterator(
columns,
dataset="%s_gene_ensembl" % query_species)
map_identifiers = collections.defaultdict(set)
for row in data:
ensid = row['ensembl_gene_id']
for column in columns[1:]:
xid = str(row[column])
if xid == "NA" or xid == "":
continue
if ensid.startswith('LRG_'):
continue
map_identifiers[xid].add((ensid, column))
# convert to lists
map_identifiers = dict([(x, list(y)) for x, y in map_identifiers.items()])
return map_identifiers
def buildOrthologyMap(query_species,
target_species,
filter_type="ortholog_one2one"):
'''build map of genes in query species to
those in target species.'''
columns = ('ensembl_gene_id', '%s_homolog_ensembl_gene' % target_species,
'%s_homolog_orthology_type' % target_species,
'%s_homolog_orthology_confidence' % target_species)
data = Biomart.biomart_iterator(columns,
dataset="%s_gene_ensembl" % query_species)
map_query2target = dict([
(x['ensembl_gene_id'], x['%s_homolog_ensembl_gene' % target_species])
for x in data
if x['%s_homolog_orthology_type' % target_species] == filter_type
])
return map_query2target
def buildOrthologyMap(query_species,
target_species,
filter_type="ortholog_one2one"):
'''build map of genes in query species to
those in target species.'''
columns = ('ensembl_gene_id',
'%s_homolog_ensembl_gene' % target_species,
'%s_homolog_orthology_type' % target_species,
'%s_homolog_orthology_confidence' % target_species)
data = Biomart.biomart_iterator(
columns,
dataset="%s_gene_ensembl" % query_species)
map_query2target = dict([(
x['ensembl_gene_id'],
x['%s_homolog_ensembl_gene' % target_species]) for x in data
if x['%s_homolog_orthology_type' % target_species] == filter_type])
return map_query2target
def importKEGGAssignments(outfile, mart, host, biomart_dataset):
'''import the KEGG annotations from the R KEGG.db annotations
package. Note that since KEGG is no longer publically availible,
this is not up-to-date and maybe removed from bioconductor in
future releases
'''
if not re.match("rnorvegicus|scerevisiae|hsapiens|mmusculus",
biomart_dataset):
E.warn("KEGG.db doesn't map Entrez ids for %s, %s will"
" likely be empty" % (biomart_dataset, outfile))
R.library("KEGG.db")
E.info("getting entrez to ensembl mapping ...")
entrez2ensembl = Biomart.biomart_iterator(
("ensembl_gene_id", "entrezgene"),
biomart=mart,
dataset=biomart_dataset,
host=host,
path="/biomart/martservice")
entrez2ensembl = dict((x['entrezgene'],
x['ensembl_gene_id'])
for x in entrez2ensembl)
E.info("Done")
E.info("getting entrez to kegg mapping ... ")
entrez2path = R('as.list(KEGGEXTID2PATHID)')
E.info("Done")
E.info("Getting KEGG names")
pathnames = R('as.list(KEGGPATHID2NAME)')
pathid2name = dict(zip(pathnames.names, R.unlist(pathnames)))
E.info("Done")
outf = IOTools.openFile(outfile, "w")
outf.write("ontology\tgene_id\tkegg_ID\tkegg_name\tevidence\n")
# rx2 did not work in rpy2 2.4.2 - workaround uses
# absolute indices
for gene_column, gene in enumerate(entrez2path.names):
try:
gene = int(gene)
except ValueError:
continue
if gene in entrez2ensembl:
ensid = entrez2ensembl[gene]
else:
continue
for pathway in entrez2path[gene_column]:
pathid = re.match("[a-z]+([0-9]+)", pathway).groups()[0]
pathname = pathid2name[pathid]
outf.write(
"\t".join(["kegg", ensid, str(pathway),
pathname, "NA"]) + "\n")
def importKEGGAssignments(outfile, mart, host, biomart_dataset):
'''import the KEGG annotations from the R KEGG.db annotations
package.
.. note::
Since KEGG is no longer publically available, this is not
up-to-date and maybe removed from bioconductor in future
releases
The table written to outfile has the following columns:
``ontology``, ``gene_id``, ``kegg_ID``, ``kegg_name``,
``evidence``.
Arguments
---------
outfile : string
Output filename in :term:`tsv` format.
mart : string
Name of the biomart
host : string
Host name of the biomart server
biomart_dataset : string
Biomart dataset
'''
if not re.match("rnorvegicus|scerevisiae|hsapiens|mmusculus",
biomart_dataset):
E.warn("KEGG.db doesn't map Entrez ids for %s, %s will"
" likely be empty" % (biomart_dataset, outfile))
R.library("KEGG.db")
E.info("getting entrez to ensembl mapping ...")
# Generates an iterator containing the data from biomart
entrez2ensembl = Biomart.biomart_iterator(
("ensembl_gene_id", "entrezgene"),
biomart=mart,
dataset=biomart_dataset,
host=host,
path="/biomart/martservice")
entrez2ensembl = dict((x['entrezgene'],
x['ensembl_gene_id'])
for x in entrez2ensembl)
E.info("Done")
E.info("getting entrez to kegg mapping ... ")
entrez2path = R('as.list(KEGGEXTID2PATHID)')
E.info("Done")
E.info("Getting KEGG names")
pathnames = R('as.list(KEGGPATHID2NAME)')
pathid2name = dict(list(zip(pathnames.names, R.unlist(pathnames))))
E.info("Done")
outf = IOTools.openFile(outfile, "w")
outf.write("ontology\tgene_id\tkegg_ID\tkegg_name\tevidence\n")
# rx2 did not work in rpy2 2.4.2 - workaround uses
# absolute indices
for gene_column, gene in enumerate(entrez2path.names):
try:
gene = int(gene)
except ValueError:
continue
if gene in entrez2ensembl:
ensid = entrez2ensembl[gene]
else:
continue
for pathway in entrez2path[gene_column]:
pathid = re.match("[a-z]+([0-9]+)", pathway).groups()[0]
pathname = pathid2name[pathid]
outf.write(
"\t".join(["kegg", ensid, str(pathway),
pathname, "NA"]) + "\n")
