def ncbi(args: argparse.Namespace) -> None:
gff = GFF.parse(args.infile).break_bubbles()
so = Ontology.from_obo_library(args.so)
name_to_so = {term.name: term for term in so.values()}
add_so_as_ontologies(gff, name_to_so)
add_ncrna_types(gff, name_to_so, so, NCRNA_TYPES)
add_pseudogene_types(gff, name_to_so, so, PSEUDOGENE_TYPES)
return
def ontology_from_obo_library(ontology_short_name: str) -> Ontology:
"""parse an ontology
This is taken directly from pronto Ontology right
now. Parse an OBO, JSON-graph, or OWL format ontology.
Parameters
==========
ontology_short_name: str
The short name from (cl.obo for cell line, ncit.obo for NCIT, etc.)
Return
======
An pronto Ontology object
"""
ont = Ontology.from_obo_library(ontology_short_name)
return ont
from pronto import Ontology
cl = Ontology("http://purl.obolibrary.org/obo/cl.obo")
#for ARGs (ARO for antibiotic resistance ontology--aro.owl from https://card.mcmaster.ca/download)
aro = Ontology.from_obo_library("aro.owl")
#exploring ontology
cf = aro['confers_resistance_to_antibiotic']
t = aro['ARO:1000001']
list(t.objects(cf))
list(t.superclasses())
list(t.subclasses())
#change format to obo
with open('aro.obo', 'wb') as f:
aro.dump(f, format='obo')
#find terms
aro = Ontology("aro.obo")
for term in aro.terms():
if term.is_leaf():
print(term.id)
#load resfinder sequences for matching
import pandas as pd
resfinder_seq = pd.read_csv("resfinder.csv", sep=" ", header=None)
aro2seq = {}
for a in set(resfinder_seq['#Aminoglycoside']):
if a not in aro:
continue
t = aro[a]
