def task_get_ssn(enzyme_type, score, hide_mutants, only_biocatdb):
job = get_current_job()
job.meta['progress'] = 'started'
job.save_meta()
ssn = SSN(enzyme_type)
job.meta['progress'] = 'ssn loaded'
job.save_meta()
if only_biocatdb is False and str(score) in ssn.db_object.precalculated_vis:
nodes = ssn.db_object.precalculated_vis[str(score)]
# Need to filter mutants here
else:
ssn.load(include_mutants=not hide_mutants, only_biocatdb=only_biocatdb)
vis = SSN_Visualiser(enzyme_type, log_level=1)
nodes, edges = vis.visualise(ssn, score)
edges = []
result = {'nodes': nodes,
'edges': edges,
'alignment_score': score,
'enzyme_type': enzyme_type,
'hide_mutants': hide_mutants,
'only_biocatdb': only_biocatdb}
return result
def get_clusters():
enzyme_type = request.form['enzyme_type']
alignment_score = int(request.form['alignment_score'])
only_biocatdb = bool(strtobool(request.form['only_biocatdb']))
hide_mutants = bool(strtobool(request.form['hide_mutants']))
ssn = SSN(enzyme_type)
ssn.load(include_mutants=not hide_mutants, only_biocatdb=only_biocatdb)
without_uniref, with_uniref = ssn.get_clusters(alignment_score)
result = {'with_uniref': with_uniref, 'without_uniref': without_uniref}
return jsonify(result=result)
def precalculate_job(enzyme_type):
ssn = SSN(enzyme_type)
ssn.load(include_mutants=False, only_biocatdb=False)
ssn_precalc = SSN_Cluster_Precalculator(ssn)
num_nodes = len(list(ssn.graph.nodes))
if num_nodes > 3000:
num = 1
elif num_nodes > 1000:
num = 5
else:
num = 20
if len(list(ssn.db_object.precalculated_vis.keys())) == 0:
print('No existing % identity data, starting at alignment score 40')
ssn_precalc.start = 40
current_num_clusters = 0
else:
start_list = [
int(s)
for s in list(ssn.db_object.identity_at_alignment_score.keys())
]
current_num_clusters = max(
list(ssn.db_object.num_at_alignment_score.values()))
ssn_precalc.start = max(start_list) + 5
print(f"Start = {ssn_precalc.start}")
precalculated_nodes, cluster_numbers, identity_at_score = ssn_precalc.precalulate(
num=num, current_num_clusters=current_num_clusters)
if len(precalculated_nodes) == 0:
print('Precalc complete - checking SSN again')
current_app.alignment_queue.enqueue(task_expand_ssn, enzyme_type)
else:
ssn.db_object.precalculated_vis.update(precalculated_nodes)
ssn.db_object.num_at_alignment_score.update(cluster_numbers)
ssn.db_object.identity_at_alignment_score.update(identity_at_score)
ssn.db_object.save()
current_app.preprocess_queue.enqueue(precalculate_job, enzyme_type)
def remove_sequence(enzyme_type, enzyme_name):
ssn = SSN(enzyme_type)
ssn.load()
if len(list(ssn.graph.nodes)) != 0:
if enzyme_name in list(ssn.graph.nodes):
ssn.graph.nodes.remove(enzyme_name)
ssn.save()
current_app.alignment_queue.enqueue(task_expand_ssn, enzyme_type)
def new_expand_ssn_job(enzyme_type):
ssn = SSN(enzyme_type)
if ssn.db_object.status != 'Complete':
current_app.alignment_queue.enqueue(task_expand_ssn, enzyme_type)
def task_expand_ssn(enzyme_type, log_level=1, max_num=200):
current_app.app_context().push()
aba_blaster = AllByAllBlaster(enzyme_type, log_level=log_level)
aba_blaster.make_blast_db()
ssn = SSN(enzyme_type, aba_blaster=aba_blaster, log_level=log_level)
ssn.load()
ssn.set_status('Checking SSN')
ssn.remove_nonexisting_seqs()
ssn.remove_seqs_marked_with_no_alignments()
biocatdb_seqs = ssn.nodes_not_present(only_biocatdb=True, max_num=max_num)
if len(biocatdb_seqs) != 0:
ssn.clear_position_information()
ssn.set_status('Adding and aligning BioCatDB sequences')
ssn.add_multiple_proteins(biocatdb_seqs)
ssn.save()
current_app.alignment_queue.enqueue(new_expand_ssn_job, enzyme_type)
return
need_alignments = ssn.nodes_need_alignments(max_num=max_num)
if len(need_alignments) != 0:
ssn.clear_position_information()
ssn.set_status('Aligning sequences in SSN')
ssn.add_multiple_proteins(need_alignments)
ssn.save()
current_app.alignment_queue.enqueue(new_expand_ssn_job, enzyme_type)
return
not_present = ssn.nodes_not_present(max_num=max_num)
if len(not_present) != 0:
ssn.clear_position_information()
ssn.set_status('Adding UniRef sequences which are not yet present')
ssn.add_multiple_proteins(not_present)
ssn.save()
current_app.alignment_queue.enqueue(new_expand_ssn_job, enzyme_type)
return
if ssn.db_object.precalculated_vis == {} and len(ssn.graph.nodes) >= 20:
ssn.set_status('Precalculating visualisations')
current_app.preprocess_queue.enqueue(precalculate_job, enzyme_type)
else:
ssn.set_status('Complete')
print(f'- SSN CONSTRUCTION FOR {enzyme_type} IS COMPLETE -')
ssn.db_object.save()