def get_gene_pmids(genes):
pmids = []
for gene in genes:
pmids_gene = pubmed_client.get_ids_for_gene(gene)
print('%s: %d' % (gene, len(pmids_gene)))
pmids += pmids_gene
return list(set(pmids))
def get_pmids(gene_names):
pmids = []
for gene_name in gene_names:
pm = get_ids_for_gene(gene_name)
pmids += pm
print('%s: %d PMIDs' % (gene_name, len(pm)))
return pmids
def get_pmids(ambig_terms):
term_pmids = {}
pmid_counter = Counter()
for term in ambig_terms:
key = (term.db, term.id)
if term.db == 'HGNC':
gene = term.entry_name
try:
term_pmids[key] = pubmed_client.get_ids_for_gene(gene)
except ValueError:
print('Could not get PMIDs for gene: %s' % gene)
term_pmids[key] = []
pmid_counter.update(term_pmids[key])
time.sleep(0.5)
elif term.db == 'MESH':
pmids = pubmed_client.get_ids_for_mesh(term.id, major_topic=False)
if len(pmids) > 1000:
pmids = pubmed_client.get_ids_for_mesh(term.id,
major_topic=True)
term_pmids[key] = pmids[:1000]
pmid_counter.update(term_pmids[key])
time.sleep(0.5)
else:
print('Unhandled ambiguous term: %s' % str(key))
term_pmids = {
k: [p for p in pmids if pmid_counter[p] == 1]
for k, pmids in term_pmids.items()
}
return term_pmids
def get_gene_pmids(gene_names):
"""Return PMIDs for all genes of interest."""
genes_pmid_list = []
for gene in gene_names:
genes_pmid_list += pubmed_client.get_ids_for_gene(gene)
genes_pmid_list = list(set(genes_pmid_list))
print('Found %d PMIDs for genes' % len(genes_pmid_list))
return genes_pmid_list
def get_gene_pmids(genes):
all_pmids = set()
for gene in genes:
print(gene)
pmids = pubmed_client.get_ids_for_gene(gene)
all_pmids = all_pmids.union(set(pmids))
all_pmids = sorted(list(all_pmids))
return all_pmids
def get_searchgenes_pmids(search_genes, num_days):
pmids = {}
for s in search_genes:
try:
pmids[s] = pubmed_client.get_ids_for_gene(s, reldate=num_days)
except ValueError as e:
logger.error('Gene symbol %s is invalid')
continue
return pmids
def get_searchgenes_pmids(search_genes, num_days):
pmids = {}
for s in search_genes:
try:
pmids[s] = pubmed_client.get_ids_for_gene(s, reldate=num_days)
except ValueError:
logger.error('Gene symbol %s is invalid')
continue
return pmids
def get_gene_pmids(genes, out_file='pmids.txt'):
all_pmids = set()
for gene in genes:
print(gene)
pmids = pubmed_client.get_ids_for_gene(gene)
all_pmids = all_pmids.union(set(pmids))
all_pmids = sorted(list(all_pmids))
with open(out_file, 'wb') as fh:
for pmid in all_pmids:
fh.write(('%s\n' % pmid).encode('utf-8'))
return all_pmids
def get_gene_pmids(genes, out_file='pmids.txt'):
all_pmids = set()
for gene in genes:
print(gene)
pmids = pubmed_client.get_ids_for_gene(gene)
all_pmids = all_pmids.union(set(pmids))
all_pmids = sorted(list(all_pmids))
with open(out_file, 'wb') as fh:
for pmid in all_pmids:
fh.write(('%s\n' % pmid).encode('utf-8'))
return all_pmids
def get_text_content_for_gene(hgnc_name):
"""Get articles that have been annotated to contain gene in entrez
Parameters
----------
hgnc_name : str
HGNC name for gene
Returns
-------
text_content : list of str
xmls of fulltext if available otherwise abstracts for all articles
that haven been annotated in entrez to contain the given gene
"""
pmids = pubmed_client.get_ids_for_gene(hgnc_name)
return get_text_content_for_pmids(pmids)
def get_text_content_for_gene(hgnc_name):
"""Get articles that have been annotated to contain gene in entrez
Parameters
----------
hgnc_name : str
HGNC name for gene
Returns
-------
text_content : list of str
xmls of fulltext if available otherwise abstracts for all articles
that haven been annotated in entrez to contain the given gene
"""
pmids = pubmed_client.get_ids_for_gene(hgnc_name)
return get_text_content_for_pmids(pmids)
def get_ids():
"""Search PubMed for references for the Ras 227 gene set."""
# Check if we've got the files already
if os.path.isfile('reading/pmids.pkl') and \
os.path.isfile('reading/pmids_from_gene.pkl'):
with open('reading/pmids.pkl') as pmids_file:
pmids = pickle.load(pmids_file)
with open('reading/pmids_from_gene.pkl') as pmids_from_gene_file:
pmids_from_gene = pickle.load(pmids_from_gene_file)
return (pmids, pmids_from_gene)
# STEP 0: Get gene list
gene_list = []
# Get gene list from ras_pathway_proteins.csv
fname = os.path.join(indra.__path__[0], 'resources',
'ras_pathway_proteins.csv')
with open(fname) as f:
csvreader = csv.reader(f, delimiter='\t')
for row in csvreader:
gene_list.append(row[0].strip())
pmids = OrderedDict()
pmids_from_gene = OrderedDict()
for gene in gene_list:
print("Querying for %s" % gene)
ids_gene = set(pubmed_client.get_ids_for_gene(gene))
print("Found %d in gene query" % len(ids_gene))
# Hack to deal with excessive number of names
if gene == 'MET':
query_gene = 'CMET'
elif gene == 'JUN':
query_gene = 'CJUN'
else:
query_gene = gene
ids_pubmed = set(
pubmed_client.get_ids(query_gene, **{'retmax': 100000}))
print("Found %d in string query" % len(ids_pubmed))
pmids[gene] = ids_pubmed
pmids_from_gene[gene] = ids_gene
with open('reading/pmids.pkl', 'wb') as f:
pickle.dump(pmids, f)
with open('reading/pmids_from_gene.pkl', 'wb') as f:
pickle.dump(pmids_from_gene, f)
return pmids, pmids_from_gene
def get_ids():
"""Search PubMed for references for the Ras 227 gene set."""
# Check if we've got the files already
if os.path.isfile('reading/pmids.pkl') and \
os.path.isfile('reading/pmids_from_gene.pkl'):
with open('reading/pmids.pkl') as pmids_file:
pmids = pickle.load(pmids_file)
with open('reading/pmids_from_gene.pkl') as pmids_from_gene_file:
pmids_from_gene = pickle.load(pmids_from_gene_file)
return (pmids, pmids_from_gene)
# STEP 0: Get gene list
gene_list = []
# Get gene list from ras_pathway_proteins.csv
with open('../../data/ras_pathway_proteins.csv') as f:
csvreader = csv.reader(f, delimiter='\t')
for row in csvreader:
gene_list.append(row[0].strip())
pmids = OrderedDict()
pmids_from_gene = OrderedDict()
for gene in gene_list:
print "Querying for", gene
ids_gene = set(pubmed_client.get_ids_for_gene(gene))
print "Found %d in gene query" % len(ids_gene)
# Hack to deal with excessive number of names
if gene == 'MET':
query_gene = 'CMET'
elif gene == 'JUN':
query_gene = 'CJUN'
else:
query_gene = gene
ids_pubmed = set(pubmed_client.get_ids(query_gene,
**{'retmax': 100000}))
print "Found %d in string query" % len(ids_pubmed)
pmids[gene] = ids_pubmed
pmids_from_gene[gene] = ids_gene
with open('reading/pmids.pkl', 'w') as f:
pickle.dump(pmids, f)
with open('reading/pmids_from_gene.pkl', 'w') as f:
pickle.dump(pmids_from_gene, f)
return (pmids, pmids_from_gene)
def test_get_ids_for_gene():
ids = pubmed_client.get_ids_for_gene('EXOC1')
assert ids
assert unicode_strs(ids)
def get_pmids_entrez(kinase):
pmids = get_ids_for_gene(kinase)
time.sleep(1)
return pmids
import pickle
from indra.literature import pubmed_client
from indra.tools.reading import submit_reading_pipeline as sub_aws
from indra.tools import assemble_corpus as ac
from indra.util import write_unicode_csv
basename = sys.argv[1]
# Get gene list
with open('genes.txt', 'rt') as f:
genes = [line.strip() for line in f.readlines()]
# Assemble a list of PMIDs curated in Entrez gene
pmids_for_genes = {}
for gene_ix, gene in enumerate(genes):
try:
pmids = pubmed_client.get_ids_for_gene(gene)
except ValueError:
print("%s: Invalid gene name, skipping" % gene)
continue
print("%s: %d articles" % (gene, len(pmids)))
pmids_for_genes[gene] = pmids
pmids = set(
[pmid for pmid_list in pmids_for_genes.values() for pmid in pmid_list])
# Save the PMIDs to a file
print("Saving PMIDs")
with open('lab_meeting_pmids.txt', 'wt') as f:
for pmid in pmids:
f.write('%s\n' % pmid)
#job_ids = sub_aws.submit_run_reach(basename, 'lab_meeting_pmids.txt',
dict_filename = 'pmids_for_gene.pkl'
if os.path.exists(dict_filename):
with open(dict_filename, 'rb') as f:
pmids_for_gene = pickle.load(f)
else:
pmids_for_gene = {}
# Get PMIDs for each HGNC ID
num_added = 0
for hgnc_name in hgnc_names:
# If HGN
#print('Getting PMIDs for %s' % hgnc_name)
if hgnc_name in pmids_for_gene:
print('%s: already got PMIDs, skipping' % hgnc_name)
continue
try:
pmids = pubmed_client.get_ids_for_gene(hgnc_name)
except ValueError as ex:
print("Exception in gettting PMIDs for %s: %s" % (hgnc_name, ex))
print("Continuing...")
continue
print('%s: %d PMIDs' % (hgnc_name, len(pmids)))
pmids_for_gene[hgnc_name] = pmids
num_added += 1
if num_added % 50 == 0:
print("Saving info for %d genes" % len(pmids_for_gene))
with open(dict_filename, 'wb') as f:
pickle.dump(pmids_for_gene, f)
unique_pmids = set([
pmid for pmid_list in pmids_for_gene.values()
for pmid in pmid_list
])
def test_get_ids_for_gene():
ids = pubmed_client.get_ids_for_gene('EXOC1')
assert ids
assert unicode_strs(ids)
def test_get_ids_for_gene():
time.sleep(0.3)
ids = pubmed_client.get_ids_for_gene('EXOC1')
assert ids
assert unicode_strs(ids)
def test_get_ids_for_gene():
time.sleep(0.5)
ids = pubmed_client.get_ids_for_gene('EXOC1')
assert ids
def test_get_ids_for_gene():
time.sleep(0.3)
ids = pubmed_client.get_ids_for_gene('EXOC1')
assert ids
assert unicode_strs(ids)
