def get_relevant_signor_statements():
"""Get Inhibition and Activation statements from SIGNOR
Returns
----------
signor_stmts_by_id : dict
Dictionary mapping SIGNOR IDs to lists of associated statements
"""
print('Collecting SIGNOR statements from the web.')
sp = signor.process_from_web()
signor_stmts_by_id = defaultdict(list)
for stmt in sp.statements:
if isinstance(stmt, Inhibition) or isinstance(stmt, Activation):
signor_stmts_by_id[stmt.evidence[0].source_id].append(stmt)
signor_stmts_by_id = dict(signor_stmts_by_id)
return signor_stmts_by_id
def _dump_test_data(filename, num_per_type=10):
"""Get corpus of statements for testing that has a range of stmt types."""
sp = signor.process_from_web()
# Group statements by type
stmts_by_type = defaultdict(list)
for stmt in sp.statements:
stmts_by_type[stmt.__class__].append(stmt)
# Sample statements of each type (without replacement)
stmt_sample = []
for stmt_type, stmt_list in stmts_by_type.items():
if len(stmt_list) <= num_per_type:
stmt_sample.extend(stmt_list)
else:
stmt_sample.extend(random.sample(stmt_list, num_per_type))
# Make a random binary class vector for the stmt list
y_arr = [random.choice((0, 1)) for s in stmt_sample]
with open(test_stmt_path, 'wb') as f:
pickle.dump((stmt_sample, y_arr), f)
return stmt_sample
def _get_statements(self):
from indra.sources.signor import process_from_web
proc = process_from_web()
return proc.statements
def get_signor_stmts():
"""Return a list of activity and a list of amount regulation stmts."""
sp = signor.process_from_web()
return ac.filter_by_type(sp.statements, RegulateActivity), \
ac.filter_by_type(sp.statements, RegulateAmount)
import sys
import pickle
from indra.sources import signor
from .util import get_mod_sites
if __name__ == '__main__':
output_file = sys.argv[1]
sp = signor.process_from_web()
sites = get_mod_sites(sp.statements)
with open(output_file, 'wb') as f:
pickle.dump(sites, f)