def test_act_phos_to_af():
act_st = Activation(Agent('A', activity=ActivityCondition('kinase', True)),
Agent('B'))
phos_st = Phosphorylation(Agent('A'), Agent('B'))
ml = MechLinker([act_st, phos_st])
linked_stmts = ml.link_statements()
assert(len(linked_stmts) == 1)
def test_replace_complexes():
phos = Phosphorylation(Agent('b'), Agent('a'))
cplx = Complex([Agent('a'), Agent('b')])
ml = MechLinker([phos, cplx])
ml.replace_complexes()
assert len(ml.statements) == 1
print(ml.statements)
def test_replace_complexes():
phos = Phosphorylation(Agent('b'), Agent('a'))
cplx = Complex([Agent('a'), Agent('b')])
ml = MechLinker([phos, cplx])
ml.replace_complexes()
assert len(ml.statements) == 1
print(ml.statements)
def test_act_phos_to_af():
act_st = Activation(Agent('A', activity=ActivityCondition('kinase', True)),
Agent('B'))
phos_st = Phosphorylation(Agent('A'), Agent('B'))
ml = MechLinker([act_st, phos_st])
linked_stmts = ml.link_statements()
assert (len(linked_stmts) == 1)
def test_replace_activations():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
phos = Phosphorylation(Agent('b'), Agent('a'))
act = Activation(Agent('b'), Agent('a'))
ml = MechLinker([af, phos, act])
ml.replace_activations()
assert len(ml.statements) == 2
print(ml.statements)
def test_act_af_to_phos():
act_st = Activation(Agent('A', activity=ActivityCondition('kinase', True)),
Agent('B'))
af_st = ActiveForm(Agent('B', mods=[ModCondition('phosphorylation',
None, None, True)]),
'activity', True)
ml = MechLinker([act_st, af_st])
linked_stmts = ml.infer_modifications(ml.statements)
assert len(linked_stmts) == 1
def test_act_af_to_phos():
act_st = Activation(Agent('A', activity=ActivityCondition('kinase', True)),
Agent('B'))
af_st = ActiveForm(
Agent('B', mods=[ModCondition('phosphorylation', None, None, True)]),
'activity', True)
ml = MechLinker([act_st, af_st])
linked_stmts = ml.infer_modifications(ml.statements)
assert len(linked_stmts) == 1
def test_replace_activations():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
phos = Phosphorylation(Agent('b'), Agent('a'))
act = Activation(Agent('b'), Agent('a'))
ml = MechLinker([af, phos, act])
ml.replace_activations()
assert len(ml.statements) == 2
print(ml.statements)
def test_require_active_forms_mod1():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
ph = Phosphorylation(Agent('a'), Agent('b'))
ml = MechLinker([af, ph])
ml.gather_explicit_activities()
ml.require_active_forms()
assert len(ml.statements) == 2
assert ml.statements[1].enz.mods
def test_reduce_mods1():
phos1 = Phosphorylation(Agent('b'), Agent('a'))
phos2 = Phosphorylation(Agent('c'), Agent('a'), 'T')
phos3 = Phosphorylation(Agent('d'), Agent('a'), 'T', '143')
ml = MechLinker([phos1, phos2, phos3])
ml.gather_modifications()
ml.reduce_modifications()
assert len(ml.statements) == 3
for st in ml.statements:
assert st.residue == 'T'
assert st.position == '143'
def test_require_active_forms_mod4():
mc1 = ModCondition('phosphorylation', 'T', '185')
mc2 = ModCondition('phosphorylation', 'Y', '187')
af = ActiveForm(Agent('a', mods=[mc1, mc2]),
'kinase', True)
ph = Phosphorylation(Agent('a', mods=[mc1]), Agent('b'))
ml = MechLinker([af, ph])
ml.gather_explicit_activities()
ml.require_active_forms()
assert len(ml.statements) == 2
assert len(ml.statements[1].enz.mods) == 2
def test_reduce_activity_types():
a1 = Agent('a', location='cytoplasm')
a2 = Agent('a', location='nucleus')
af1 = ActiveForm(a1, 'activity', True)
af2 = ActiveForm(a2, 'kinase', True)
af3 = ActiveForm(a1, 'catalytic',True)
ml = MechLinker([af1, af2, af3])
ml.gather_explicit_activities()
ml.reduce_activities()
assert af1.activity == 'kinase'
assert af2.activity == 'kinase'
assert af3.activity == 'kinase'
def test_infer_activations():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
phos = Phosphorylation(Agent('b'), Agent('a'))
linked_stmts = MechLinker.infer_activations([af, phos])
assert len(linked_stmts) == 1
print(linked_stmts)
def test_infer_activations():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
phos = Phosphorylation(Agent('b'), Agent('a'))
linked_stmts = MechLinker.infer_activations([af, phos])
assert len(linked_stmts) == 1
print(linked_stmts)
def assemble_model(requester_name):
global stmts
# Performing grounding mapping on the statements
gmapper = gm.GroundingMapper(gm.default_grounding_map)
stmts = gmapper.map_agents(stmts)
pa = Preassembler(hierarchies, stmts)
pa.combine_related()
stmts = pa.related_stmts
ml = MechLinker(stmts)
linked_stmts = ml.link_statements()
if linked_stmts:
for linked_stmt in linked_stmts:
if linked_stmt.inferred_stmt:
question = mechlinker_queries.print_linked_stmt(linked_stmt)
say(question)
stmts.append(linked_stmt.inferred_stmt)
say("%s: Done, updating layout." % requester_name)
update_layout()
def test_reduce_mods2():
mc1 = ModCondition('phosphorylation', 'S', '123', False)
mc2 = ModCondition('phosphorylation', 'S', None, True)
mc3 = ModCondition('phosphorylation', 'T')
mc4 = ModCondition('phosphorylation', 'T', '111')
mc5 = ModCondition('phosphorylation', 'T', '999')
mc6 = ModCondition('phosphorylation')
mc7 = ModCondition('phosphorylation', None, '999')
st1 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc1]))
st2 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc2]))
st3 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc3]))
st4 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc4]))
st5 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc5]))
st6 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc6]))
st7 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc7]))
ml = MechLinker([st1, st2, st3, st4, st5, st6, st7])
ml.gather_modifications()
ml.reduce_modifications()
assert len(ml.statements) == 7
mc_red1 = ml.statements[0].obj.mods[0]
mc_red2 = ml.statements[1].obj.mods[0]
mc_red3 = ml.statements[2].obj.mods[0]
mc_red4 = ml.statements[3].obj.mods[0]
mc_red5 = ml.statements[4].obj.mods[0]
mc_red6 = ml.statements[5].obj.mods[0]
mc_red7 = ml.statements[6].obj.mods[0]
# These ones stay the same because they shouldn't be reduced
assert mc_red1.__dict__ == mc1.__dict__
assert mc_red3.__dict__ == mc3.__dict__
assert mc_red4.__dict__ == mc4.__dict__
assert mc_red5.__dict__ == mc5.__dict__
assert mc_red6.__dict__ == mc6.__dict__
# mc2 has to be reduced to have position '123'
assert mc_red2.mod_type == 'phosphorylation'
assert mc_red2.residue == 'S'
assert mc_red2.position == '123'
assert mc_red2.is_modified == True
# mc7 has to be reduced to have residue 'T'
assert mc_red7.mod_type == 'phosphorylation'
assert mc_red7.residue == 'T'
assert mc_red7.position == '999'
assert mc_red7.is_modified == True
def reduce_activities(stmts_in, **kwargs):
"""Reduce the activity types in a list of statements
Parameters
----------
stmts_in : list[indra.statements.Statement]
A list of statements to reduce activity types in.
save : Optional[str]
The name of a pickle file to save the results (stmts_out) into.
Returns
-------
stmts_out : list[indra.statements.Statement]
A list of reduced activity statements.
"""
logger.info('Reducing activities on %d statements...' % len(stmts_in))
stmts_out = [deepcopy(st) for st in stmts_in]
ml = MechLinker(stmts_out)
ml.get_activities()
ml.reduce_activities()
stmts_out = ml.statements
dump_pkl = kwargs.get('save')
if dump_pkl:
dump_statements(stmts_out, dump_pkl)
return stmts_out
def get_missing_activities(self):
ml = MechLinker(self.statements)
ml.gather_explicit_activities()
suggestions = []
for stmt in self.statements:
if isinstance(stmt,
(Modification, RegulateActivity, RegulateAmount)):
# The subj here is in an "active" position
subj, obj = stmt.agent_list()
if subj is None:
continue
subj_base = ml._get_base(subj)
# If it has any activities but isn't in an active state
# here
if subj_base.activity_types and not subj.activity:
# We suggest making the subj active in this case
suggestion = deepcopy(stmt)
act_type = subj_base.activity_types[0]
new_subj = deepcopy(subj)
new_subj.activity = ActivityCondition(act_type, True)
suggestion.set_agent_list([new_subj, obj])
suggestions.append(suggestion)
return suggestions
def get_missing_activities(self):
ml = MechLinker(self.statements)
ml.gather_explicit_activities()
suggestions = []
for stmt in self.statements:
if isinstance(stmt, (Modification, RegulateActivity,
RegulateAmount)):
# The subj here is in an "active" position
subj, obj = stmt.agent_list()
if subj is None:
continue
subj_base = ml._get_base(subj)
# If it has any activities but isn't in an active state
# here
if subj_base.activity_types and not subj.activity:
# We suggest making the subj active in this case
suggestion = deepcopy(stmt)
act_type = subj_base.activity_types[0]
new_subj = deepcopy(subj)
new_subj.activity = ActivityCondition(act_type, True)
suggestion.set_agent_list([new_subj, obj])
suggestions.append(suggestion)
return suggestions
def test_require_active_forms_act1():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
act = Activation(Agent('a'), Agent('b'))
ml = MechLinker([af, act])
ml.gather_explicit_activities()
ml.require_active_forms()
assert (len(ml.statements) == 2)
assert (ml.statements[1].subj.mods)
def test_require_active_forms_mod1():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
ph = Phosphorylation(Agent('a'), Agent('b'))
ml = MechLinker([af, ph])
ml.gather_explicit_activities()
ml.require_active_forms()
assert len(ml.statements) == 2
assert ml.statements[1].enz.mods
def test_require_active_forms_mod2():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
af2 = ActiveForm(Agent('a', location='nucleus'), 'activity', True)
ph = Phosphorylation(Agent('a'), Agent('b'))
ml = MechLinker([af, af2, ph])
ml.gather_explicit_activities()
ml.require_active_forms()
assert (len(ml.statements) == 4)
assert (ml.statements[3].enz.location)
def test_require_active_forms_mod4():
mc1 = ModCondition('phosphorylation', 'T', '185')
mc2 = ModCondition('phosphorylation', 'Y', '187')
af = ActiveForm(Agent('a', mods=[mc1, mc2]), 'kinase', True)
ph = Phosphorylation(Agent('a', mods=[mc1]), Agent('b'))
ml = MechLinker([af, ph])
ml.gather_explicit_activities()
ml.require_active_forms()
assert len(ml.statements) == 2
assert len(ml.statements[1].enz.mods) == 2
def test_reduce_mods1():
phos1 = Phosphorylation(Agent('b'), Agent('a'))
phos2 = Phosphorylation(Agent('c'), Agent('a'), 'T')
phos3 = Phosphorylation(Agent('d'), Agent('a'), 'T', '143')
ml = MechLinker([phos1, phos2, phos3])
ml.gather_modifications()
ml.reduce_modifications()
assert len(ml.statements) == 3
for st in ml.statements:
assert st.residue == 'T'
assert st.position == '143'
def test_reduce_activity_types():
a1 = Agent('a', location='cytoplasm')
a2 = Agent('a', location='nucleus')
af1 = ActiveForm(a1, 'activity', True)
af2 = ActiveForm(a2, 'kinase', True)
af3 = ActiveForm(a1, 'catalytic', True)
ml = MechLinker([af1, af2, af3])
ml.gather_explicit_activities()
ml.reduce_activities()
assert af1.activity == 'kinase'
assert af2.activity == 'kinase'
assert af3.activity == 'kinase'
def test_reduce_mods2():
mc1 = ModCondition('phosphorylation', 'S', '123', False)
mc2 = ModCondition('phosphorylation', 'S', None, True)
mc3 = ModCondition('phosphorylation', 'T')
mc4 = ModCondition('phosphorylation', 'T', '111')
mc5 = ModCondition('phosphorylation', 'T', '999')
mc6 = ModCondition('phosphorylation')
mc7 = ModCondition('phosphorylation', None, '999')
st1 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc1]))
st2 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc2]))
st3 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc3]))
st4 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc4]))
st5 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc5]))
st6 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc6]))
st7 = Activation(Agent('KRAS'), Agent('BRAF', mods=[mc7]))
ml = MechLinker([st1, st2, st3, st4, st5, st6, st7])
ml.gather_modifications()
ml.reduce_modifications()
assert len(ml.statements) == 7
mc_red1 = ml.statements[0].obj.mods[0]
mc_red2 = ml.statements[1].obj.mods[0]
mc_red3 = ml.statements[2].obj.mods[0]
mc_red4 = ml.statements[3].obj.mods[0]
mc_red5 = ml.statements[4].obj.mods[0]
mc_red6 = ml.statements[5].obj.mods[0]
mc_red7 = ml.statements[6].obj.mods[0]
# These ones stay the same because they shouldn't be reduced
assert mc_red1.__dict__ == mc1.__dict__
assert mc_red3.__dict__ == mc3.__dict__
assert mc_red4.__dict__ == mc4.__dict__
assert mc_red5.__dict__ == mc5.__dict__
assert mc_red6.__dict__ == mc6.__dict__
# mc2 has to be reduced to have position '123'
assert mc_red2.mod_type == 'phosphorylation'
assert mc_red2.residue == 'S'
assert mc_red2.position == '123'
assert mc_red2.is_modified == True
# mc7 has to be reduced to have residue 'T'
assert mc_red7.mod_type == 'phosphorylation'
assert mc_red7.residue == 'T'
assert mc_red7.position == '999'
assert mc_red7.is_modified == True
def test_infer_complexes():
phos = Phosphorylation(Agent('b'), Agent('a'))
linked_stmts = MechLinker.infer_complexes([phos])
assert len(linked_stmts) == 1
print(linked_stmts)
def test_base_agent():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
ml = MechLinker([af])
ml.gather_explicit_activities()
def test_infer_complexes():
phos = Phosphorylation(Agent('b'), Agent('a'))
linked_stmts = MechLinker.infer_complexes([phos])
assert len(linked_stmts) == 1
print(linked_stmts)
from indra.mechlinker import MechLinker
from indra.assemblers import EnglishAssembler
def print_linked_stmt(stmt):
source_txts = []
for source_stmt in stmt.source_stmts:
source_txt = EnglishAssembler([source_stmt]).make_model()
source_txts.append(source_txt)
query_txt = EnglishAssembler([stmt.inferred_stmt]).make_model()
final_txt = 'I know that '
for i, t in enumerate(source_txts):
final_txt += '(%d) %s ' % (i + 1, t)
if i < len(source_txts) - 1:
final_txt = final_txt[:-2] + ', and '
final_txt += 'Is it therefore true that ' + query_txt[:-1] + '?'
print(final_txt)
return final_txt
if __name__ == '__main__':
fname = 'models/rasmachine/rem/model.pkl'
model = IncrementalModel(fname)
model.preassemble()
stmts = model.assembled_stmts
linked_stmts = MechLinker.infer_active_forms(stmts)
linked_stmts += MechLinker.infer_modifications(stmts)
linked_stmts += MechLinker.infer_activations(stmts)
for stmt in linked_stmts:
print_linked_stmt(stmt)
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
from indra.mechlinker import MechLinker
from indra.assemblers.english import EnglishAssembler
def print_linked_stmt(stmt):
source_txts = []
for source_stmt in stmt.source_stmts:
source_txt = EnglishAssembler([source_stmt]).make_model()
source_txts.append(source_txt)
query_txt = EnglishAssembler([stmt.inferred_stmt]).make_model()
final_txt = 'I know that '
for i, t in enumerate(source_txts):
final_txt += '(%d) %s ' % (i+1, t)
if i < len(source_txts) -1:
final_txt = final_txt[:-2] + ', and '
final_txt += 'Is it therefore true that ' + query_txt[:-1] + '?'
print(final_txt)
return final_txt
if __name__ == '__main__':
fname = 'models/rasmachine/rem/model.pkl'
model = IncrementalModel(fname)
model.preassemble()
stmts = model.assembled_stmts
linked_stmts = MechLinker.infer_active_forms(stmts)
linked_stmts += MechLinker.infer_modifications(stmts)
linked_stmts += MechLinker.infer_activations(stmts)
for stmt in linked_stmts:
print_linked_stmt(stmt)
def run_assembly(stmts, folder, pmcid, background_assertions=None):
'''Run assembly on a list of statements, for a given PMCID.'''
# Folder for index card output (scored submission)
indexcard_prefix = folder + '/index_cards/' + pmcid
# Folder for other outputs (for analysis, debugging)
otherout_prefix = folder + '/other_outputs/' + pmcid
# Do grounding mapping here
# Load the TRIPS-specific grounding map and add to the default
# (REACH-oriented) grounding map:
trips_gm = load_grounding_map('trips_grounding_map.csv')
default_grounding_map.update(trips_gm)
gm = GroundingMapper(default_grounding_map)
mapped_agent_stmts = gm.map_agents(stmts)
renamed_agent_stmts = gm.rename_agents(mapped_agent_stmts)
# Filter for grounding
grounded_stmts = []
for st in renamed_agent_stmts:
if all([is_protein_or_chemical(a) for a in st.agent_list()]):
grounded_stmts.append(st)
# Instantiate the Preassembler
pa = Preassembler(hierarchies)
pa.add_statements(grounded_stmts)
print('== %s ====================' % pmcid)
print('%d statements collected in total.' % len(pa.stmts))
# Combine duplicates
unique_stmts = pa.combine_duplicates()
print('%d statements after combining duplicates.' % len(unique_stmts))
# Run BeliefEngine on unique statements
epe = BeliefEngine()
epe.set_prior_probs(pa.unique_stmts)
# Build statement hierarchy
related_stmts = pa.combine_related()
# Run BeliefEngine on hierarchy
epe.set_hierarchy_probs(related_stmts)
print('%d statements after combining related.' % len(related_stmts))
# Instantiate the mechanism linker
ml = MechLinker(related_stmts)
# Link statements
linked_stmts = ml.link_statements()
# Run BeliefEngine on linked statements
epe.set_linked_probs(linked_stmts)
# Print linked statements for debugging purposes
print('Linked\n=====')
for ls in linked_stmts:
print(ls.inferred_stmt.belief, ls.inferred_stmt)
print('=============')
# Combine all statements including linked ones
all_statements = ml.statements + [ls.inferred_stmt for ls in linked_stmts]
# Instantiate a new preassembler
pa = Preassembler(hierarchies, all_statements)
# Build hierarchy again
pa.combine_duplicates()
# Choose the top-level statements
related_stmts = pa.combine_related()
# Remove top-level statements that came only from the prior
if background_assertions is not None:
nonbg_stmts = [
stmt for stmt in related_stmts if stmt not in background_assertions
]
else:
nonbg_stmts = related_stmts
# Dump top-level statements in a pickle
with open(otherout_prefix + '.pkl', 'wb') as fh:
pickle.dump(nonbg_stmts, fh, protocol=2)
# Flatten evidence for statements
flattened_evidence_stmts = flatten_evidence(nonbg_stmts)
# Start a card counter
card_counter = 1
# We don't limit the number of cards reported in this round
card_lim = float('inf')
top_stmts = []
###############################################
# The belief cutoff for statements
belief_cutoff = 0.3
###############################################
# Sort by amount of evidence
for st in sorted(flattened_evidence_stmts,
key=lambda x: x.belief,
reverse=True):
if st.belief >= belief_cutoff:
print(st.belief, st)
if st.belief < belief_cutoff:
print('SKIP', st.belief, st)
# If it's background knowledge, we skip the statement
if is_background_knowledge(st):
print('This statement is background knowledge - skipping.')
continue
# Assemble IndexCards
ia = IndexCardAssembler([st], pmc_override=pmcid)
ia.make_model()
# If the index card was actually made
# (not all statements can be assembled into index cards to
# this is often not the case)
if ia.cards:
# Save the index card json
ia.save_model(indexcard_prefix + '-%d.json' % card_counter)
card_counter += 1
top_stmts.append(st)
if card_counter > card_lim:
break
# Print the English-assembled model for debugging purposes
ea = EnglishAssembler(top_stmts)
print('=======================')
print(ea.make_model())
print('=======================')
# Print the statement graph
graph = render_stmt_graph(nonbg_stmts)
graph.draw(otherout_prefix + '_graph.pdf', prog='dot')
# Print statement diagnostics
print_stmts(pa.stmts, otherout_prefix + '_statements.tsv')
print_stmts(related_stmts, otherout_prefix + '_related_statements.tsv')
def assemble_pysb(stmts, data_genes, contextualize=False):
# Filter the INDRA Statements to be put into the model
stmts = ac.filter_by_type(stmts, Complex, invert=True)
stmts = ac.filter_direct(stmts)
stmts = ac.filter_belief(stmts, 0.95)
stmts = ac.filter_top_level(stmts)
# Strip the extraneous supports/supported by here
strip_supports(stmts)
stmts = ac.filter_gene_list(stmts, data_genes, 'all')
stmts = ac.filter_enzyme_kinase(stmts)
stmts = ac.filter_mod_nokinase(stmts)
stmts = ac.filter_transcription_factor(stmts)
# Simplify activity types
ml = MechLinker(stmts)
ml.gather_explicit_activities()
ml.reduce_activities()
ml.gather_modifications()
ml.reduce_modifications()
stmts = normalize_active_forms(ml.statements)
# Replace activations when possible
ml = MechLinker(stmts)
ml.gather_explicit_activities()
ml.replace_activations()
# Require active forms
ml.require_active_forms()
num_stmts = len(ml.statements)
while True:
# Remove inconsequential PTMs
ml.statements = ac.filter_inconsequential_mods(ml.statements,
get_mod_whitelist())
ml.statements = ac.filter_inconsequential_acts(ml.statements,
get_mod_whitelist())
if num_stmts <= len(ml.statements):
break
num_stmts = len(ml.statements)
stmts = ml.statements
# Save the Statements here
ac.dump_statements(stmts, prefixed_pkl('pysb_stmts'))
# Add drug target Statements
drug_target_stmts = get_drug_target_statements()
stmts += drug_target_stmts
# Just generate the generic model
pa = PysbAssembler()
pa.add_statements(stmts)
model = pa.make_model()
with open(prefixed_pkl('pysb_model'), 'wb') as f:
pickle.dump(model, f)
# Run this extra part only if contextualize is set to True
if not contextualize:
return
cell_lines_no_data = ['COLO858', 'K2', 'MMACSF', 'MZ7MEL', 'WM1552C']
for cell_line in cell_lines:
if cell_line not in cell_lines_no_data:
stmtsc = contextualize_stmts(stmts, cell_line, data_genes)
else:
stmtsc = stmts
pa = PysbAssembler()
pa.add_statements(stmtsc)
model = pa.make_model()
if cell_line not in cell_lines_no_data:
contextualize_model(model, cell_line, data_genes)
ac.dump_statements(stmtsc, prefixed_pkl('pysb_stmts_%s' % cell_line))
with open(prefixed_pkl('pysb_model_%s' % cell_line), 'wb') as f:
pickle.dump(model, f)
def run_assembly(stmts, folder, pmcid):
indexcard_prefix = folder + '/index_cards/' + pmcid
otherout_prefix = folder + '/other_outputs/' + pmcid
# Filter for grounding
grounded_stmts = []
for st in stmts:
if all([is_protein_or_chemical(a) for a in st.agent_list()]):
grounded_stmts.append(st)
# Instantiate the Preassembler
pa = Preassembler(eh, mh)
pa.add_statements(grounded_stmts)
print '%d statements collected in total.' % len(pa.stmts)
unique_stmts = pa.combine_duplicates()
print '%d statements after combining duplicates.' % len(unique_stmts)
ml = MechLinker(unique_stmts)
ml.link_statements()
pa = Preassembler(eh, mh, ml.statements)
pa.combine_duplicates()
related_stmts = pa.combine_related()
print '%d statements after combining related.' % len(related_stmts)
with open(otherout_prefix + '.pkl', 'wb') as fh:
pickle.dump(related_stmts, fh)
flattened_evidence_stmts = flatten_evidence(related_stmts)
card_counter = 1
card_lim = float('inf')
top_stmts = []
for st in sorted(flattened_evidence_stmts,
key=lambda x: len(x.evidence), reverse=True):
print len(st.evidence), st
if is_background_knowledge(st):
print 'This statement is background knowledge - skipping.'
continue
# Assemble IndexCards
ia = IndexCardAssembler([st])
ia.make_model()
if ia.cards:
ia.save_model(indexcard_prefix + '-%d.json' % card_counter)
card_counter += 1
top_stmts.append(st)
if card_counter > card_lim:
break
ea = EnglishAssembler(top_stmts)
print '======================='
print ea.make_model()
print '======================='
# Print the statement graph
graph = render_stmt_graph(related_stmts)
graph.draw(otherout_prefix + '_graph.pdf', prog='dot')
# Print statement diagnostics
print_stmts(pa.stmts, otherout_prefix + '_statements.tsv')
print_stmts(related_stmts, otherout_prefix + '_related_statements.tsv')
pya = PysbAssembler()
pya.add_statements(related_stmts)
model = pya.make_model()
print 'PySB model has %d monomers and %d rules' %\
(len(model.monomers), len(model.rules))
def run_assembly(stmts, folder, pmcid, background_assertions=None):
'''Run assembly on a list of statements, for a given PMCID.'''
# Folder for index card output (scored submission)
indexcard_prefix = folder + '/index_cards/' + pmcid
# Folder for other outputs (for analysis, debugging)
otherout_prefix = folder + '/other_outputs/' + pmcid
# Do grounding mapping here
# Load the TRIPS-specific grounding map and add to the default
# (REACH-oriented) grounding map:
trips_gm = load_grounding_map('trips_grounding_map.csv')
default_grounding_map.update(trips_gm)
gm = GroundingMapper(default_grounding_map)
mapped_agent_stmts = gm.map_agents(stmts)
renamed_agent_stmts = gm.rename_agents(mapped_agent_stmts)
# Filter for grounding
grounded_stmts = []
for st in renamed_agent_stmts:
if all([is_protein_or_chemical(a) for a in st.agent_list()]):
grounded_stmts.append(st)
# Instantiate the Preassembler
pa = Preassembler(hierarchies)
pa.add_statements(grounded_stmts)
print('== %s ====================' % pmcid)
print('%d statements collected in total.' % len(pa.stmts))
# Combine duplicates
unique_stmts = pa.combine_duplicates()
print('%d statements after combining duplicates.' % len(unique_stmts))
# Run BeliefEngine on unique statements
epe = BeliefEngine()
epe.set_prior_probs(pa.unique_stmts)
# Build statement hierarchy
related_stmts = pa.combine_related()
# Run BeliefEngine on hierarchy
epe.set_hierarchy_probs(related_stmts)
print('%d statements after combining related.' % len(related_stmts))
# Instantiate the mechanism linker
# Link statements
linked_stmts = MechLinker.infer_active_forms(related_stmts)
linked_stmts += MechLinker.infer_modifications(related_stmts)
linked_stmts += MechLinker.infer_activations(related_stmts)
# Run BeliefEngine on linked statements
epe.set_linked_probs(linked_stmts)
# Print linked statements for debugging purposes
print('Linked\n=====')
for ls in linked_stmts:
print(ls.inferred_stmt.belief, ls.inferred_stmt)
print('=============')
# Combine all statements including linked ones
all_statements = related_stmts + [ls.inferred_stmt for ls in linked_stmts]
# Instantiate a new preassembler
pa = Preassembler(hierarchies, all_statements)
# Build hierarchy again
pa.combine_duplicates()
# Choose the top-level statements
related_stmts = pa.combine_related()
# Remove top-level statements that came only from the prior
if background_assertions is not None:
nonbg_stmts = [stmt for stmt in related_stmts
if stmt not in background_assertions]
else:
nonbg_stmts = related_stmts
# Dump top-level statements in a pickle
with open(otherout_prefix + '.pkl', 'wb') as fh:
pickle.dump(nonbg_stmts, fh)
# Flatten evidence for statements
flattened_evidence_stmts = flatten_evidence(nonbg_stmts)
# Start a card counter
card_counter = 1
# We don't limit the number of cards reported in this round
card_lim = float('inf')
top_stmts = []
###############################################
# The belief cutoff for statements
belief_cutoff = 0.3
###############################################
# Sort by amount of evidence
for st in sorted(flattened_evidence_stmts,
key=lambda x: x.belief, reverse=True):
if st.belief >= belief_cutoff:
print(st.belief, st)
if st.belief < belief_cutoff:
print('SKIP', st.belief, st)
# If it's background knowledge, we skip the statement
if is_background_knowledge(st):
print('This statement is background knowledge - skipping.')
continue
# Assemble IndexCards
ia = IndexCardAssembler([st], pmc_override=pmcid)
ia.make_model()
# If the index card was actually made
# (not all statements can be assembled into index cards to
# this is often not the case)
if ia.cards:
# Save the index card json
ia.save_model(indexcard_prefix + '-%d.json' % card_counter)
card_counter += 1
top_stmts.append(st)
if card_counter > card_lim:
break
# Print the English-assembled model for debugging purposes
ea = EnglishAssembler(top_stmts)
print('=======================')
print(ea.make_model().encode('utf-8'))
print('=======================')
# Print the statement graph
graph = render_stmt_graph(nonbg_stmts)
graph.draw(otherout_prefix + '_graph.pdf', prog='dot')
# Print statement diagnostics
print_stmts(pa.stmts, otherout_prefix + '_statements.tsv')
print_stmts(related_stmts, otherout_prefix + '_related_statements.tsv')
def preprocess_stmts(stmts, data_genes):
# Filter the INDRA Statements to be put into the model
stmts = ac.filter_mutation_status(stmts,
{'BRAF': [('V', '600', 'E')]}, ['PTEN'])
stmts = ac.filter_by_type(stmts, Complex, invert=True)
stmts = ac.filter_direct(stmts)
stmts = ac.filter_belief(stmts, 0.95)
stmts = ac.filter_top_level(stmts)
stmts = ac.filter_gene_list(stmts, data_genes, 'all')
stmts = ac.filter_enzyme_kinase(stmts)
stmts = ac.filter_mod_nokinase(stmts)
stmts = ac.filter_transcription_factor(stmts)
# Simplify activity types
ml = MechLinker(stmts)
ml.gather_explicit_activities()
ml.reduce_activities()
ml.gather_modifications()
ml.reduce_modifications()
af_stmts = ac.filter_by_type(ml.statements, ActiveForm)
non_af_stmts = ac.filter_by_type(ml.statements, ActiveForm, invert=True)
af_stmts = ac.run_preassembly(af_stmts)
stmts = af_stmts + non_af_stmts
# Replace activations when possible
ml = MechLinker(stmts)
ml.gather_explicit_activities()
ml.replace_activations()
# Require active forms
ml.require_active_forms()
num_stmts = len(ml.statements)
while True:
# Remove inconsequential PTMs
ml.statements = ac.filter_inconsequential_mods(ml.statements,
get_mod_whitelist())
ml.statements = ac.filter_inconsequential_acts(ml.statements,
get_mod_whitelist())
if num_stmts <= len(ml.statements):
break
num_stmts = len(ml.statements)
stmts = ml.statements
return stmts
from indra.tools.incremental_model import IncrementalModel
from indra.mechlinker import MechLinker
from indra.assemblers import EnglishAssembler
def print_linked_stmt(stmt):
source_txts = []
for source_stmt in stmt.source_stmts:
source_txt = EnglishAssembler([source_stmt]).make_model()
source_txts.append(source_txt)
query_txt = EnglishAssembler([stmt.inferred_stmt]).make_model()
final_txt = 'I know that '
for i, t in enumerate(source_txts):
final_txt += '(%d) %s ' % (i + 1, t)
if i < len(source_txts) - 1:
final_txt = final_txt[:-2] + ', and '
final_txt += 'Is it therefore true that ' + query_txt[:-1] + '?'
print(final_txt)
return final_txt
if __name__ == '__main__':
fname = 'models/rasmachine/rem/model.pkl'
model = IncrementalModel(fname)
model.preassemble()
stmts = model.toplevel_stmts
ml = MechLinker(stmts)
linked_stmts = ml.link_statements()
for stmt in linked_stmts:
print_linked_stmt(stmt)
import pickle
from indra.tools.incremental_model import IncrementalModel
from indra.mechlinker import MechLinker
from indra.assemblers import EnglishAssembler
def print_linked_stmt(stmt):
source_txts = []
for source_stmt in stmt.source_stmts:
source_txt = EnglishAssembler([source_stmt]).make_model()
source_txts.append(source_txt)
query_txt = EnglishAssembler([stmt.inferred_stmt]).make_model()
final_txt = 'I know that '
for i, t in enumerate(source_txts):
final_txt += '(%d) %s ' % (i+1, t)
if i < len(source_txts) -1:
final_txt = final_txt[:-2] + ', and '
final_txt += 'Is it therefore true that ' + query_txt[:-1] + '?'
print(final_txt)
return final_txt
if __name__ == '__main__':
fname = 'models/rasmachine/rem/model.pkl'
model = IncrementalModel(fname)
model.preassemble()
stmts = model.toplevel_stmts
ml = MechLinker(stmts)
linked_stmts = ml.link_statements()
for stmt in linked_stmts:
print_linked_stmt(stmt)
def test_base_agent():
af = ActiveForm(Agent('a', mods=[ModCondition('phosphorylation')]),
'activity', True)
ml = MechLinker([af])
ml.gather_explicit_activities()
