def test_merge_deltas():
def add_annots(stmt):
for ev in stmt.evidence:
ev.annotations['subj_adjectives'] = stmt.subj_delta['adjectives']
ev.annotations['obj_adjectives'] = stmt.obj_delta['adjectives']
ev.annotations['subj_polarity'] = stmt.subj_delta['polarity']
ev.annotations['obj_polarity'] = stmt.obj_delta['polarity']
return stmt
d1 = {'adjectives': ['a', 'b', 'c'], 'polarity': 1}
d2 = {'adjectives': [], 'polarity': -1}
d3 = {'adjectives': ['g'], 'polarity': 1}
d4 = {'adjectives': ['d', 'e', 'f'], 'polarity': -1}
d5 = {'adjectives': ['d'], 'polarity': None}
d6 = {'adjectives': [], 'polarity': None}
d7 = {'adjectives': [], 'polarity': 1}
stmts = [
add_annots(
Influence(Concept('a'),
Concept('b'),
subj_delta=sd,
obj_delta=od,
evidence=[Evidence(source_api='eidos',
text='%d' % idx)]))
for idx, (sd, od) in enumerate([(d1, d2), (d3, d4)])
]
stmts = ac.run_preassembly(stmts, return_toplevel=True)
stmts = ac.merge_deltas(stmts)
assert stmts[0].subj_delta['polarity'] == 1, stmts[0].subj_delta
assert stmts[0].obj_delta['polarity'] == -1, stmts[0].obj_delta
assert set(stmts[0].subj_delta['adjectives']) == {'a', 'b', 'c', 'g'}, \
stmts[0].subj_delta
assert set(stmts[0].obj_delta['adjectives']) == {'d', 'e', 'f'}, \
stmts[0].obj_delta
stmts = [
add_annots(
Influence(Concept('a'),
Concept('b'),
subj_delta=sd,
obj_delta=od,
evidence=[Evidence(source_api='eidos',
text='%d' % idx)]))
for idx, (sd, od) in enumerate([(d1, d5), (d6, d7), (d6, d7)])
]
stmts = ac.run_preassembly(stmts, return_toplevel=True)
stmts = ac.merge_deltas(stmts)
assert stmts[0].subj_delta['polarity'] is None, stmts[0].subj_delta
assert stmts[0].obj_delta['polarity'] == 1, stmts[0].obj_delta
assert set(stmts[0].subj_delta['adjectives']) == {'a', 'b', 'c'}, \
stmts[0].subj_delta
assert set(stmts[0].obj_delta['adjectives']) == {'d'}, \
stmts[0].obj_delta
def test_merge_deltas():
def add_annots(stmt):
for ev in stmt.evidence:
ev.annotations['subj_adjectives'] = stmt.subj.delta.adjectives
ev.annotations['obj_adjectives'] = stmt.obj.delta.adjectives
ev.annotations['subj_polarity'] = stmt.subj.delta.polarity
ev.annotations['obj_polarity'] = stmt.obj.delta.polarity
return stmt
# d1 = {'adjectives': ['a', 'b', 'c'], 'polarity': 1}
# d2 = {'adjectives': [], 'polarity': -1}
# d3 = {'adjectives': ['g'], 'polarity': 1}
# d4 = {'adjectives': ['d', 'e', 'f'], 'polarity': -1}
# d5 = {'adjectives': ['d'], 'polarity': None}
# d6 = {'adjectives': [], 'polarity': None}
# d7 = {'adjectives': [], 'polarity': 1}
d1 = QualitativeDelta(polarity=1, adjectives=['a', 'b', 'c'])
d2 = QualitativeDelta(polarity=-1, adjectives=None)
d3 = QualitativeDelta(polarity=1, adjectives=['g'])
d4 = QualitativeDelta(polarity=-1, adjectives=['d', 'e', 'f'])
d5 = QualitativeDelta(polarity=None, adjectives=['d'])
d6 = QualitativeDelta(polarity=None, adjectives=None)
d7 = QualitativeDelta(polarity=1, adjectives=None)
def make_ev(name, delta):
return Event(Concept(name), delta=delta)
stmts = [add_annots(Influence(make_ev('a', sd), make_ev('b', od),
evidence=[Evidence(source_api='eidos',
text='%d' % idx)]))
for idx, (sd, od) in enumerate([(d1, d2), (d3, d4)])]
stmts = ac.run_preassembly(stmts, return_toplevel=True)
stmts = ac.merge_deltas(stmts)
assert stmts[0].subj.delta.polarity == 1, stmts[0].subj.delta
assert stmts[0].obj.delta.polarity == -1, stmts[0].obj.delta
assert set(stmts[0].subj.delta.adjectives) == {'a', 'b', 'c', 'g'}, \
stmts[0].subj.delta
assert set(stmts[0].obj.delta.adjectives) == {'d', 'e', 'f'}, \
stmts[0].obj.delta
stmts = [add_annots(Influence(make_ev('a', sd), make_ev('b', od),
evidence=[Evidence(source_api='eidos',
text='%d' % idx)]))
for idx, (sd, od) in enumerate([(d1, d5), (d6, d7), (d6, d7)])]
stmts = ac.run_preassembly(stmts, return_toplevel=True)
stmts = ac.merge_deltas(stmts)
assert stmts[0].subj.delta.polarity is None, stmts[0].subj.delta
assert stmts[0].obj.delta.polarity == 1, stmts[0].obj.delta
assert set(stmts[0].subj.delta.adjectives) == {'a', 'b', 'c'}, \
stmts[0].subj.delta
assert set(stmts[0].obj.delta.adjectives) == {'d'}, \
stmts[0].obj.delta
def run_assembly(self):
"""Run INDRA's assembly pipeline on the Statements."""
self.eliminate_copies()
stmts = self.get_indra_stmts()
stmts = self.filter_event_association(stmts)
stmts = ac.filter_no_hypothesis(stmts)
if not self.assembly_config.get('skip_map_grounding'):
stmts = ac.map_grounding(stmts)
if self.assembly_config.get('standardize_names'):
ac.standardize_names_groundings(stmts)
if self.assembly_config.get('filter_ungrounded'):
score_threshold = self.assembly_config.get('score_threshold')
stmts = ac.filter_grounded_only(stmts,
score_threshold=score_threshold)
if self.assembly_config.get('merge_groundings'):
stmts = ac.merge_groundings(stmts)
if self.assembly_config.get('merge_deltas'):
stmts = ac.merge_deltas(stmts)
relevance_policy = self.assembly_config.get('filter_relevance')
if relevance_policy:
stmts = self.filter_relevance(stmts, relevance_policy)
if not self.assembly_config.get('skip_filter_human'):
stmts = ac.filter_human_only(stmts)
if not self.assembly_config.get('skip_map_sequence'):
stmts = ac.map_sequence(stmts)
# Use WM hierarchies and belief scorer for WM preassembly
preassembly_mode = self.assembly_config.get('preassembly_mode')
if preassembly_mode == 'wm':
hierarchies = get_wm_hierarchies()
belief_scorer = get_eidos_scorer()
stmts = ac.run_preassembly(stmts,
return_toplevel=False,
belief_scorer=belief_scorer,
hierarchies=hierarchies)
else:
stmts = ac.run_preassembly(stmts, return_toplevel=False)
belief_cutoff = self.assembly_config.get('belief_cutoff')
if belief_cutoff is not None:
stmts = ac.filter_belief(stmts, belief_cutoff)
stmts = ac.filter_top_level(stmts)
if self.assembly_config.get('filter_direct'):
stmts = ac.filter_direct(stmts)
stmts = ac.filter_enzyme_kinase(stmts)
stmts = ac.filter_mod_nokinase(stmts)
stmts = ac.filter_transcription_factor(stmts)
if self.assembly_config.get('mechanism_linking'):
ml = MechLinker(stmts)
ml.gather_explicit_activities()
ml.reduce_activities()
ml.gather_modifications()
ml.reduce_modifications()
ml.gather_explicit_activities()
ml.replace_activations()
ml.require_active_forms()
stmts = ml.statements
self.assembled_stmts = stmts
def default_assembly(stmts):
from indra.belief.wm_scorer import get_eidos_scorer
from indra.preassembler.hierarchy_manager import get_wm_hierarchies
hm = get_wm_hierarchies()
scorer = get_eidos_scorer()
stmts = ac.run_preassembly(stmts, belief_scorer=scorer,
return_toplevel=True,
flatten_evidence=True,
flatten_evidence_collect_from='supported_by',
poolsize=4)
stmts = ac.merge_groundings(stmts)
stmts = ac.merge_deltas(stmts)
stmts = ac.standardize_names_groundings(stmts)
return stmts
def default_assembly(stmts):
from indra.belief.wm_scorer import get_eidos_scorer
from indra.preassembler.hierarchy_manager import get_wm_hierarchies
hm = get_wm_hierarchies()
scorer = get_eidos_scorer()
stmts = ac.run_preassembly(stmts, belief_scorer=scorer,
return_toplevel=True,
flatten_evidence=True,
flatten_evidence_collect_from='supported_by',
poolsize=4)
stmts = ac.merge_groundings(stmts)
stmts = ac.merge_deltas(stmts)
stmts = ac.standardize_names_groundings(stmts)
return stmts
def surface_details(stmts):
stmts = ac.merge_groundings(stmts)
stmts = ac.merge_deltas(stmts)
return stmts
from indra.statements import stmts_to_json_file
import indra.tools.assemble_corpus as ac
from assemble_model import process_eidos, assemble_stmts
if __name__ == '__main__':
stmts = process_eidos()
stmts_to_json_file(stmts, 'eidos_500m_raw.json')
stmts = assemble_stmts(stmts)
stmts = ac.merge_groundings(stmts)
stmts = ac.merge_deltas(stmts)
stmts = ac.standardize_names_groundings(stmts)
stmts_to_json_file(stmts, 'eidos_500m_assembled.json')