#!/usr/bin/env python3
import csv
from os.path import expanduser
import rdflib
from pyontutils.scigraph_client import Vocabulary, Graph
from IPython import embed
dbx = 'http://www.geneontology.org/formats/oboInOwl#hasDbXref'
with open(expanduser('~/git/entity_mapping/mappings/uberon-nervous'),
'rt') as f:
brain_only = set([l.strip() for l in f.readlines()])
v = Vocabulary('http://localhost:9000/scigraph')
sg = Graph('http://localhost:9000/scigraph')
g = rdflib.Graph()
g.parse(expanduser('~/git/NIF-Ontology/ttl/generated/cocomacslim.ttl'),
format='turtle')
sos = [so for so in g.subject_objects(rdflib.RDFS.label)]
map_ = []
smap_ = []
fma_lookup = {}
for s, o in sos:
cc_id = g.qname(s)
cc_label = o.toPython()
existing_id = None
existing_label = None
existing_fma = ''
def main():
sgg = Graph(cache=True)
sgg_local = Graph('http://localhost:9000/scigraph', cache=True)
cell = Query("GO:0044464", 'subClassOf', 'INCOMING', 9)
nifga = Query('NIFGA:birnlex_796',
'http://www.obofoundry.org/ro/ro.owl#has_proper_part',
'OUTGOING', 9)
uberon = Query('UBERON:0000955',
'http://purl.obolibrary.org/obo/BFO_0000050', 'INCOMING', 9)
uberon_ae = Query('UBERON:0001062', 'subClassOf', 'INCOMING',
9) # anatomical entity
uberon_cc = Query('UBERON:0002749',
'http://purl.obolibrary.org/obo/BFO_0000050', 'INCOMING',
9)
ncbi_ins = Query('NCBITaxon:50557', 'subClassOf', 'INCOMING', 10)
ncbi_rod = Query('NCBITaxon:9989', 'subClassOf', 'INCOMING', 10)
queries = cell, nifga, uberon, uberon_cc
queries = ncbi_ins, ncbi_rod, uberon, nifga
queries = uberon_ae,
fma3_r = Query('FMA3:Brain',
'http://sig.biostr.washington.edu/fma3.0#regional_part_of',
'INCOMING', 9)
fma3_c = Query(
'FMA3:Brain',
'http://sig.biostr.washington.edu/fma3.0#constitutional_part_of',
'INCOMING', 9)
#fma3_tree, fma3_extra = creatTree(*fma3_r, graph=sgg_local)
fma_r = Query('FMA:50801', 'http://purl.org/sig/ont/fma/regional_part_of',
'INCOMING', 20)
fma_c = Query('FMA:50801',
'http://purl.org/sig/ont/fma/constitutional_part_of',
'INCOMING', 20)
fma_rch_r = Query('FMA:61819',
'http://purl.org/sig/ont/fma/regional_part_of',
'INCOMING', 20)
#fma_tree, fma_extra = creatTree(*fma_r, graph=sgg_local)
#fma_tree, fma_extra = creatTree(*fma_rch_r, graph=sgg_local)
fma_hip = Query('FMA:275020',
'http://purl.org/sig/ont/fma/regional_part_of', 'BOTH', 20)
fma_hip = Query('FMA:275020',
'http://purl.org/sig/ont/fma/constitutional_part_of',
'BOTH', 20)
#fma_tree, fma_extra = creatTree(*fma_hip, graph=sgg_local)
fma_mfg = Query('FMA:273103',
'http://purl.org/sig/ont/fma/regional_part_of', 'BOTH', 20)
#fma_tree, fma_extra = creatTree(*fma_mfg, graph=sgg_local)
ncbi_metazoa = Query('NCBITaxon:33208', 'subClassOf', 'INCOMING', 20)
ncbi_vertebrata = Query('NCBITaxon:7742', 'subClassOf', 'INCOMING', 40)
#ncbi_tree, ncbi_extra = creatTree(*ncbi_vertebrata, graph=sgg_local)
fma_tel = Query('FMA:62000',
'http://purl.org/sig/ont/fma/regional_part_of', 'INCOMING',
20)
DOFMA = False
if DOFMA:
fma_gsc_tree, fma_gsc_extra = creatTree(*fma_tel, graph=sgg_local)
childs = list(
fma_gsc_extra[2]
) # get the curies for the left/right so we can get parents for all
from heatmaps.scigraph_client import Graph
g = Graph('http://localhost:9000/scigraph')
parent_nodes = []
for curie in childs:
json = g.getNeighbors(curie, relationshipType='subClassOf')
if json:
for node in json['nodes']:
if node['id'] != curie:
parent_nodes.append(node) # should have dupes
embed()
return
uberon_tree, uberon_extra = creatTree(
*uberon,
graph=sgg) # ,url_base='localhost:9000') # FIXME why does this crash?
print(uberon_tree)
print(uberon_extra[0])
embed()
return
uberon_flat = [n.replace(':', '_') for n in flatten(uberon_extra[0])]
with open('/tmp/uberon_partonomy_terms', 'wt') as f:
f.writelines('\n'.join(uberon_flat))
for query in queries:
tree, extra = creatTree(*query, graph=sgg)
dematerialize(list(tree.keys())[0], tree)
print(tree)
#print(extra[0])
with open('/tmp/' + query.root, 'wt') as f:
f.writelines(tree.print_tree())
level_sizes = [len(levels(tree, i)) for i in range(11)]
print('level sizes', level_sizes)
parent_counts = np.unique([len(v) for v in extra[-4].values()])
print('unique parent counts', parent_counts)
print('num terms', len(extra[2]))
embed()
