def get_pmid_from_url(url):
if anyMembers(url,
'www.ncbi.nlm.nih.gov/pubmed/',
'europepmc.org/abstract/MED/'):
# TODO validate the suffix
_, suffix = url.rsplit('/', 1)
return 'PMID:' + suffix
def main():
from docopt import docopt
args = docopt(__doc__, version='ont-catalog 0.0.1')
dobig = args['--big']
remote_base = 'http://ontology.neuinfo.org/NIF/ttl/'
olr = Path(args['--ontology-local-repo'])
local_base = (olr / 'ttl').as_posix() + '/'
#list of all nif ontologies
#onts = [f for f in fs if f.endswith('.ttl') or f.endswith('.owl') and 'NEMO_' not in f]
repo = Repo(olr)
repo_path = Path(olr)
tracked_files = [
(repo_path / f).as_posix()
# FIXME missing scicrunch-registry.ttl
for f in repo.git.ls_files('--', 'ttl/').split('\n')
if f.endswith('.ttl') or f.endswith('.owl')
]
#_ = [print(f) for f in fs]
extra_files = [] # TODO pass in via cli?
mapping = [(remote_base + fragment, fragment)
for file in tracked_files + extra_files
for _, fragment in (file.split('/ttl/', 1), )]
# check for mismatched import and ontology iris
itrips = local_imports(
remote_base, local_base, tracked_files, readonly=True,
dobig=dobig) # XXX these files are big and slow, run at own peril
sa = {os.path.basename(o): s for s, p, o in itrips if 'sameAs' in p}
# FIXME should be able to do this by checking what is tracked by git...
externals = ('CogPO.owl', 'NEMO_vv2.98.owl', 'cogat_v0.3.owl', 'doid.owl',
'ero.owl', 'pato.owl', 'pr.owl', 'ro_bfo1-1_bridge.owl',
'uberon.owl')
for f in tracked_files + extra_files:
if '/external/' in f and anyMembers(f, *externals):
basename = os.path.basename(f)
if basename in sa:
target = sa[basename]
if 'external' not in target:
mapping.append((target, 'external/' + basename))
# make a protege catalog file to simplify life
uriline = ' <uri id="User Entered Import Resolution" name="{ontid}" uri="{filename}"/>'
xmllines = ['<?xml version="1.0" encoding="UTF-8" standalone="no"?>',
'<catalog prefer="public" xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog">',] + \
[uriline.format(ontid=ont, filename=file) for ont,file in sorted(mapping)] + \
['</catalog>']
xml = '\n'.join(xmllines)
with open(f'{tempfile.tempdir}/nif-catalog-v001.xml', 'wt') as f:
f.write(xml)
if args['--debug']:
breakpoint()
def extract(og, ng, curie, mkdir=False):
graph = og.g
properties = (owl.AnnotationProperty, owl.DatatypeProperty,
owl.ObjectProperty)
deads = [s for s in graph.subjects(owl.deprecated, rdflib.Literal(True))]
for s in deads:
types = set(o for o in graph.objects(s, rdf.type))
if anyMembers(types, *properties):
p, o = rdfs.subPropertyOf, owl.DeprecatedProperty
elif owl.Class in types:
p, o = rdfs.subClassOf, owl.DeprecatedClass
else:
continue # don't bother with named individuals
trip = (ng.check_thing(s), ng.check_thing(p), ng.check_thing(o))
if trip not in og.g:
ng.g.add(trip)
# TODO cases where owl:deprecated is not used but sco owl:DeprecatedClass is...
base_alts = list(graph.subject_objects(oboInOwl.hasAlternativeId))
for replacedByClass, oldClassString in base_alts:
if curie + ':' in oldClassString or curie + '_' in oldClassString:
oldClassString = oldClassString.toPython()
s = ng.check_thing(oldClassString)
if s not in deads:
types = set(o
for o in graph.objects(replacedByClass, rdf.type))
if anyMembers(types, *properties):
p, o = rdfs.subPropertyOf, owl.DeprecatedProperty
elif owl.Class in types:
p, o = rdfs.subClassOf, owl.DeprecatedClass
else:
continue # don't bother with named individuals
[ng.add_trip(s, rdf.type, o) for o in types]
ng.add_trip(s, p, o)
ng.add_trip(s, owl.deprecated, True)
ng.add_trip(s, 'replacedBy:', replacedByClass)
if mkdir:
parent = Path(ng.filename).parent
if not parent.exists():
parent.mkdir(parents=True)
ng.write()
def reverse_report():
ilx = Graph()
ilx.parse('/tmp/interlex.ttl', format='turtle')
not_in_ontology = set()
annotations = set()
relations = set()
drugbank = set()
t3db = set()
for subject in ilx.subjects(rdf.type, owl.Class):
ok = False
for object in ilx.objects(subject, oboInOwl.hasDbXref):
if anyMembers(object, 'uri.neuinfo.org', 'GO_', 'CHEBI_', 'PR_',
'PATO_', 'HP_', 'OBI_', 'DOID_', 'COGPO_', 'CAO_',
'UBERON_', 'NCBITaxon_', 'SO_', 'IAO_'):
# FIXME doe we areally import HP?
ok = True
if (subject, rdf.type, owl.AnnotationProperty) in ilx: # FIXME for troy these need to be cleared up
annotations.add(subject)
elif (subject, rdf.type, owl.ObjectProperty) in ilx:
relations.add(subject)
elif 'drugbank' in object:
drugbank.add(subject)
elif 't3db.org' in object:
t3db.add(subject)
if not ok:
not_in_ontology.add(subject)
drugbank = drugbank & not_in_ontology
t3db = t3db & not_in_ontology
annotations = annotations & not_in_ontology
relations = relations & not_in_ontology
unaccounted = not_in_ontology - drugbank - t3db - annotations - relations
report = (
f'Total {len(not_in_ontology)}\n'
f'annotations {len(annotations)}\n'
f'relations {len(relations)}\n'
f'drugbank {len(drugbank)}\n'
f't3db {len(t3db)}\n'
f'unaccounted {len(unaccounted)}\n'
)
print(report)
return (not_in_ontology, drugbank, unaccounted)
def uri_normalization(uri):
""" NOTE: this does NOT produce uris """
try:
# strip hypothesis extension prefix
if uri.startswith('chrome-extension://bjfhmglciegochdpefhhlphglcehbmek/content/web/viewer.html?file='):
junk, uri = uri.split('=', 1)
# universal fixes
no_fragment, *_frag = uri.rsplit('#', 1)
no_trailing_slash = no_fragment.rstrip('/') # annoying
_scheme, no_scheme = no_trailing_slash.split('://', 1)
# special cases
if 'frontiersin.org' in no_scheme:
# og:url on frontiers is incorrect
no_scheme = no_scheme.replace('article/', 'articles/')
elif 'fasebj.org' in no_scheme: # FIXME this one has _all_ the variants :/
no_scheme = (no_scheme
.replace('.abstract', '')
.replace('.full', '')
.replace('.pdf', '')
)
elif no_scheme.endswith('?needAccess=true'):
no_scheme = no_scheme[:-len('?needAccess=true')]
elif '?systemMessage' in no_scheme:
no_scheme, junk = no_scheme.rsplit('?systemMessage', 1)
# specific fixes
if anyMembers(no_scheme,
'acs.org',
'ahajournals.org',
'biologicalpsychiatryjournal.com',
'ebiomedicine.com',
'fasebj.org',
'frontiersin.org',
'future-science.com',
'hindawi.com',
'ieee.org',
'jclinepi.com',
'jpeds.com',
'liebertpub.com',
'mitpressjournals.org',
'molbiolcell.org',
'molmetab.com',
'neurobiologyofaging.org',
'physiology.org',
'sagepub.com',
'sciencedirect.com',
'tandfonline.com',
'theriojournal.com',
'wiley.com',):
# NOTE not all the above hit all of these
# almost all still resolve
normalized = (no_scheme
.replace('/abstract', '')
.replace('/abs', '')
.replace('/fulltext', '')
.replace('/full', '')
.replace('/pdf', ''))
#elif ('sciencedirect.com' in no_scheme):
#normalized = (no_scheme
#.replace('/abs', ''))
elif ('cell.com' in no_scheme):
normalized = (no_scheme # FIXME looks like cell uses /abstract in og:url
.replace('/abstract', '/XXX')
.replace('/fulltext', '/XXX'))
elif 'jneurosci.org' in no_scheme:
# TODO content/early -> resolution_chain(doi)
normalized = (no_scheme
.replace('.short', '')
.replace('.long', '')
.replace('.full', '')
.replace('.pdf', '')
# note .full.pdf is a thing
)
elif 'pnas.org' in no_scheme:
normalized = (no_scheme
.replace('.short', '')
.replace('.long', '')
.replace('.full', ''))
elif 'mdpi.com' in no_scheme:
normalized = (no_scheme
.replace('/htm', ''))
elif 'f1000research.com' in no_scheme:
# you should be ashamed of yourselves for being in here for this reason
normalized, *maybe_version = no_scheme.rsplit('/v', 1)
elif 'academic.oup.com' in no_scheme:
normalized, *maybesr = no_scheme.rsplit('?searchresult=', 1)
_normalized, maybe_junk = normalized.rsplit('/', 1)
numbers = '0123456789'
if (maybe_junk[0] not in numbers or # various ways to detect the human readable junk after the id
maybe_junk[-1] not in numbers or
'-' in maybe_junk or
len(maybe_junk) > 20):
normalized = _normalized
elif anyMembers(no_scheme,
'jci.org',
'nature.com'):
# cases where safe to remove query fragment
normalized, *_query = no_scheme.rsplit('?', 1)
normalized, *table_number = normalized.rsplit('/tables/', 1)
elif 'pubmed/?term=' in no_scheme and noneMembers(no_scheme, ' ', '+'):
normalized = no_scheme.replace('?term=', '')
elif 'nih.gov/pubmed/?' in no_scheme:
# FIXME scibot vs client norm?
normalized = no_scheme.replace(' ', '+')
elif 'govhttp' in no_scheme:
# lol oh dear
hrm, oops = no_scheme.split('govhttp')
ded, wat = oops.split('//', 1)
blargh, suffix = wat.split('/', 1)
normalized = hrm + 'gov/pmc/' + suffix
elif 'table/undtbl' in no_scheme:
normalized, table_number = no_scheme.rsplit('table/undtbl')
elif anyMembers(no_scheme,
'index.php?',
):
# cases where we just use hypothes.is normalization
_scheme, normalized = uri_normalize(uri).split('://') # FIXME h dependency
else:
normalized = no_scheme
'onlinelibrary.wiley.com/doi/10.1002/cne.23727?wol1URL=/doi/10.1002/cne.23727®ionCode=US-CA&identityKey=e2523300-b934-48c9-b08e-940de05d7335'
'www.jove.com/video/55441/?language=Japanese'
'www.nature.com/neuro/journal/v19/n5/full/nn.4282.html'
'www.nature.com/cr/journal/vaop/ncurrent/full/cr201669a.html'
'https://www.nature.com/articles/cr201669'
#{'www.ingentaconnect.com/content/umrsmas/bullmar/2017/00000093/00000002/art00006':
#[OntId('DOI:10.5343/bms.2016.1044'), OntId('DOI:info:doi/10.5343/bms.2016.1044')]}
# pmid extract from pmc
#<meta name="citation_pmid" content="28955177">
return normalized
except ValueError as e: # split fail
pdf_prefix = 'urn:x-pdf:'
if uri.startswith(pdf_prefix):
return uri
elif uri in bad_uris:
print('AAAAAAAAAAAAAAAAAAAAAAAAAAA', uri)
return 'THIS URI IS GARBAGE AND THIS IS ITS NORMALIZED FORM'
else:
raise TypeError(uri) from e
# check for mismatched import and ontology iris
itrips = local_imports(remote_base,
local_base,
onts,
readonly=True,
dobig=True)
sa = {os.path.basename(o): s for s, p, o in itrips if 'sameAs' in p}
# FIXME should be able to do this by checking what is tracked by git...
externals = ('CogPO.owl', 'NEMO_vv2.98.owl', 'cogat_v0.3.owl', 'doid.owl',
'ero.owl', 'pato.owl', 'pr.owl', 'ro_bfo1-1_bridge.owl',
'uberon.owl')
for f in fs:
if '/external/' in f and anyMembers(f, *externals):
basename = os.path.basename(f)
if basename in sa:
target = sa[basename]
if 'external' not in target:
mapping.append((target, 'external/' + basename))
# make a protege catalog file to simplify life
uriline = ' <uri id="User Entered Import Resolution" name="{ontid}" uri="{filename}"/>'
xmllines = ['<?xml version="1.0" encoding="UTF-8" standalone="no"?>',
'<catalog prefer="public" xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog">',] + \
[uriline.format(ontid=ont, filename=file) for ont,file in sorted(mapping)] + \
['</catalog>']
xml = '\n'.join(xmllines)
with open('/tmp/nif-catalog-v001.xml', 'wt') as f:
def already_released_or_skipped(self):
return any(anyMembers(r.tags, self.REPLY_TAG) for r in self.replies)
def swapUriSwitch(trip, ureps, fragment_prefixes):
for spo in trip:
if not isinstance(spo, rdflib.URIRef):
yield spo, None, None
continue
elif spo in ureps:
new_spo = ureps[spo]
rep = (new_spo, owl.sameAs, spo)
if 'nlx_' in new_spo:
pref = 'nlx_'
elif '/readable/' in new_spo:
pref = 'NIFRID'
else:
pref = 'NIFSTD'
yield new_spo, rep, pref
continue
elif anyMembers(
spo, # backend refactor
'BIRNLex_annotation_properties.owl#',
'OBO_annotation_properties.owl#'):
_, suffix = spo.rsplit('#', 1)
new_spo = rdflib.URIRef(
os.path.join(NIFSTDBASE, 'readable', suffix))
rep = (new_spo, owl.sameAs, spo)
pref = 'NIFRID'
yield new_spo, rep, pref
continue
try:
uri_pref, fragment = spo.rsplit('#', 1)
if '_' in fragment:
frag_pref, p_suffix = fragment.split('_', 1)
if not p_suffix[0].isdigit():
p, suffix = p_suffix.split('_', 1)
frag_pref = frag_pref + '_' + p
else:
suffix = p_suffix
frag_pref_ = frag_pref + '_'
if frag_pref_ in fragment_prefixes:
if frag_pref_ == 'nlx_sub_': pref = 'nlx_subcell_'
elif frag_pref_ == 'nif_organ_': pref = 'nlx_organ_'
else:
pref = frag_pref_ # come on branch predictor you can do it!
elif frag_pref_ == 'nlx_neuron_': # special case
rest = 'nt_'
suffix = suffix[len(rest):]
pref = frag_pref_ + rest
else:
yield spo, None, None
continue
elif 'sao' in fragment:
suffix = fragment[3:].strip('-')
pref = 'sao'
else:
yield spo, None, None
continue
new_spo = rdflib.URIRef(NIFSTDBASE + pref + suffix)
if new_spo != spo:
rep = (new_spo, owl.sameAs, spo)
else:
rep = None
print('Already converted', spo)
yield new_spo, rep, pref
except ValueError: # there was no # so do not split
yield spo, None, None
continue
def __init__(self, args, **kwargs):
kwargs = {k: None if v == 'None' else v for k, v in kwargs.items()}
self.__dict__.update(kwargs)
self._updated = False
if self.check_built:
self.build_only = True
self._host = HOST
self._user = USER
self.args = args
self.ontload_args = {
k: None if v == 'None' else v
for k, v in args.items()
} # send them all!
self.ontload_args['scigraph'] = self.services
if self.all or self.graph:
self.ontload_args['graph'] = True
self.ontload_args.update({
'imports': None,
'chain': None,
'extra': None,
'<ontologies>': []
})
mode = [
k for k, v in self.args.items()
if not k.startswith('-') and not k.startswith('<') and v
]
self.mode = mode[0] if mode else None
self.build_services_config(
) # needed to update self.graph_folder XXX hack fixme
self._init_more()
self.same_remotes = False
if self.local and self.build_only:
if self.check_built:
self.local_dispatch()
return
elif self.build_host == self.services_host and self.build_user == self.build_user:
self.same_remotes = True
# the executor is different from the remotes
if self.build_host != self._host and not self.check_built:
#self._host = self.build_host
#self._user = self.build_user
self._building = False
for name, obj in inspect.getmembers(self):
continue # TODO there is a bug here with executor/build boundaries
# i think the issues is with the way we are calling anyMembers
if inspect.ismethod(obj) and anyMembers(
name, 'config', 'services', 'graph', 'remote'):
@wraps(obj)
def mutex_on_ssh(*args,
func=obj,
**kwargs): # ah late binding hacks
if not self._building:
self._building = True
out = func(*args, **kwargs)
self._building = False
if out.startswith('('):
#out = f'"{out[1:-1]}"'
out = out[1:-1]
print('YAY FOR ONLY ONE SSH!')
return out
#return f'ssh {self._user}@{self._host} {out}'
else:
return func(*args, **kwargs)
setattr(self, name, mutex_on_ssh)
elif not self.local:
print('WARNING: all servers are equivalent to localhost '
'but you are running without --local. Did you mean to?')
elif self.build_host == self._host and self.build_user == self._user:
if not self.local:
print('WARNING: all servers are equivalent to localhost '
'but you are running without --local. Did you mean to?')
def would_you_like_to_know_more_question_mark():
# resolving differences between classes
more_ids = set((
'http://uri.neuinfo.org/nif/nifstd/readable/ChEBIid',
'http://uri.neuinfo.org/nif/nifstd/readable/GOid',
'http://uri.neuinfo.org/nif/nifstd/readable/MeshUid',
'http://uri.neuinfo.org/nif/nifstd/readable/PMID',
'http://uri.neuinfo.org/nif/nifstd/readable/UmlsCui',
'http://uri.neuinfo.org/nif/nifstd/readable/bamsID',
'http://uri.neuinfo.org/nif/nifstd/readable/bonfireID',
'http://uri.neuinfo.org/nif/nifstd/readable/cell_ontology_ID',
'http://uri.neuinfo.org/nif/nifstd/readable/definingCitationID',
'http://uri.neuinfo.org/nif/nifstd/readable/definingCitationURI',
'http://uri.neuinfo.org/nif/nifstd/readable/emapMouseStageDataID',
'http://uri.neuinfo.org/nif/nifstd/readable/emapMouseStageDiagramID',
'http://uri.neuinfo.org/nif/nifstd/readable/externalSourceId',
'http://uri.neuinfo.org/nif/nifstd/readable/externalSourceURI',
'http://uri.neuinfo.org/nif/nifstd/readable/gbifID',
'http://uri.neuinfo.org/nif/nifstd/readable/gbifTaxonKeyID',
'http://uri.neuinfo.org/nif/nifstd/readable/gene_Ontology_ID',
#'http://uri.neuinfo.org/nif/nifstd/readable/hasExternalSource',
'http://uri.neuinfo.org/nif/nifstd/readable/hasGenbankAccessionNumber',
'http://uri.neuinfo.org/nif/nifstd/readable/imsrStandardStrainName',
'http://uri.neuinfo.org/nif/nifstd/readable/isReplacedByClass',
'http://uri.neuinfo.org/nif/nifstd/readable/jaxMiceID',
'http://uri.neuinfo.org/nif/nifstd/readable/ncbiTaxID',
'http://uri.neuinfo.org/nif/nifstd/readable/neuronamesID',
'http://uri.neuinfo.org/nif/nifstd/readable/nifID',
'http://uri.neuinfo.org/nif/nifstd/readable/sao_ID',
'http://uri.neuinfo.org/nif/nifstd/readable/umls_ID',
'http://www.geneontology.org/formats/oboInOwl#id',
))
outside = []
eee = {}
resolver_not_ilx_only_but_not_in_scigraph = set() # resources.ttl
_res = Graph().parse((gitf / 'NIF-Ontology/ttl/resources.ttl').as_posix(), format='turtle')
reslookup = {uri:[l] for uri, l in _res.subject_objects(rdfs.label)}
for uri in chain(h_uris, resolver_not_ilx_only):
if 'uri.neuinfo.org' in uri:
try:
meta = sgg.getNode(uri.toPython())['nodes'][0]['meta']
asdf = {hng.qname(k):v for k, v in meta.items() if k in more_ids}
except TypeError:
resolver_not_ilx_only_but_not_in_scigraph.add(uri) # resources.ttl ;)
if uri in reslookup: # no differentia
asdf = False
else:
asdf = False
print('WTF', uri)
if asdf:
#print(uri, asdf)
eee[uri] = asdf
for l in asdf.values():
for e in l:
outside.append(e)
outside_dupes = [v for v, c in Counter(outside).most_common() if c > 1]
eee_dupes = {k:v for k, v in eee.items() if anyMembers(outside_dupes, *(e for l in v.values() for e in l))}
#for uri, meta in sorted(eee_dupes.items(), key=lambda a:sorted(a[1].values())):
#print(uri.toPython(), sorted((e.replace('PMID: ', 'PMID:'), k) for k, l in meta.items() for e in l))
# attempt to deal with label mappings
iexisting = defaultdict(set)
iiexisting = {}
for i, existing in zip(datal('ilx'), datal('iri')):
#if 'uri.neuinfo.org' in existing:
if 'interlex.org' not in existing and 'neurolex.org' not in existing:
iexisting[i].add(URIRef(existing))
iiexisting[URIRef(existing)] = i
iexisting = {**iexisting}
_ilabs = {k:l for k, l in zip(datal('ilx'), datal('label'))}
def inner(iri):
resp = sgv.findById(iri)
if resp is not None:
l = resp['labels']
else:
l = [] #_ilabs[iiexisting[iri]] + '** already in ilx **']
#print('trouble?', iri) # ilx only
return iri, l
#labs = {k:v[0] if v else '<--NO-LABEL-->' for k, v in Async()(deferred(inner)(id_) for id_ in chain(h_uris, (e for s in iexisting.values() for e in s)))}
labs = {k:v[0] if v else '<--NO-LABEL-->' for k, v in Async()(deferred(inner)(id_) for id_ in h_uris)}
ilabs = {k:l.lower() for k, l in zip(datal('ilx'), datal('label'))}
iilabs = {v:k for k, v in ilabs.items()}
assert len(ilabs) == len(iilabs)
missing_map = {k:iilabs[v.lower()] for k, v in labs.items() if v and v.lower() in iilabs} # XXX this is not valid
missing_existing = {i:[m, *iexisting[i]] for m, i in missing_map.items() if i in iexisting}
missing_equivs = {next(iter(iexisting[i])):i for m, i in missing_map.items() if i in iexisting}
eid = NIFRID.externalSourceId.toPython()
ded = owl.deprecated.toPython()
# SP: -> swissprot vs uniprot
mmr = []
proto_mmr_1_to_1 = {}
arrr = defaultdict(set)
uniprot_iuphar = set()
for uri, ilx_frag in {**missing_equivs, **missing_map}.items():
uri = URIRef(uri)
try:
meta = sgg.getNode(uri.toPython())['nodes'][0]['meta']
except TypeError:
# just ignore these, they are ilx only :/
meta = {}
if eid in meta:
src = meta[eid][0]
if src.startswith('SP:'):
src = tc.yellow(src.replace('SP:', 'http://www.uniprot.org/uniprot/'))
#elif src.startswith('IUPHAR:'):
#pass
#else:
#src = 'TODO'
elif ded in meta and meta[ded]:
src = tc.red('ded ')
else:
src = 'TODO'
val = labs[uri] if uri in labs else _ilabs[ilx_frag] + ' **'
if uri in eee:
differentia = str(eee[uri])
for v in eee[uri].values():
for e in v:
arrr[e].add(uri)
if 'SP:' in e or 'IUPHAR:' in e:
uniprot_iuphar.add(uri)
else:
differentia = ''
if uri in _ilx and uri in all_uris:
ruri = SGG[hng.qname(uri)]
ruri = tc.blue(f'{ruri:<60}')
else:
ruri = uri
ruri = f'{ruri:<60}'
v = ' '.join((f'{val:<60}',
src,
ruri,
ilxb[ilx_frag],
differentia))
mmr.append(v)
proto_mmr_1_to_1[uri] = v
src = None
arrr = {**arrr}
arrr_not_1_to_1 = {k:v for k, v in arrr.items() if len(v) > 1}
#arrr_n11_uris = set((u.toPython() for v in arrr_not_1_to_1.values() for u in v))
arrr_n11_uris = set.union(*arrr_not_1_to_1.values())
mmr_1_to_1 = {k:v for k, v in proto_mmr_1_to_1.items() if k not in arrr_n11_uris}
no_uniprot = {k:v for k, v in proto_mmr_1_to_1.items() if k not in uniprot_iuphar}
arrr_n11_text = '\n'.join(f'{k:<15} {sorted(_.toPython() for _ in v)}' for k, v in arrr_not_1_to_1.items())
mmr.sort()
mmr_text = '\n'.join(mmr)
mmr_1_to_1_text = '\n'.join(sorted(mmr_1_to_1.values()))
no_uniprot_text = '\n'.join(sorted(no_uniprot.values()))
def main():
from docopt import docopt
args = docopt(__doc__, version='rridxp 0.0.0')
print(args)
from scibot.release import get_annos, Curation, SamePMID, MultiplePMID, MultipleDOI, MPP, MPD
annos = get_annos()
[Curation(a, annos) for a in annos]
def midr():
mp = multiIssue(MultiplePMID(Curation))
md = multiIssue(MultipleDOI(Curation))
# filtering by url first removes any detectable instances of multiple dois/pmids
#mpp = multiIssue(MPP(Curation))
#mpd = multiIssue(MPD(Curation))
with open('multiple-pmids.json', 'wt') as f:
json.dump(mp, f, sort_keys=True, indent=4, cls=Encode)
with open('multiple-dois.json', 'wt') as f:
json.dump(md, f, sort_keys=True, indent=4, cls=Encode)
if args['multi-id-report']:
midr()
elif args['csv']:
substrings = args['<filter>'] # ['MGI', 'IMSR']
if substrings:
ssj = '-'.join(ss.lower() for ss in substrings) + '-'
else:
substrings = ['']
ssj = 'all-'
pmids2 = SamePMID(
set(annotation for paper in Curation._papers.values()
for rrid, annotations in paper.items()
if rrid is not None and anyMembers(rrid, *substrings)
for annotation in annotations))
now = UTCNOW()
rows = [[
'PMID', 'DOI', 'URI', 'shareLink', 'exact', 'rrid', 'public_tags'
]]
rows += sorted(
deNone(anno.pmid, anno.doi, anno.uri, anno.shareLink, anno.exact,
anno.rrid, ','.join(
[t for t in anno.public_tags if 'RRID:' not in t]))
for pmid, papers in pmids2.items() for rrids in papers.values()
for annos in rrids.values() for anno in annos)
with open(f'{ssj}rrids-{now}.csv', 'wt') as f:
csv.writer(f, lineterminator='\n').writerows(rows)
nomatch = [[
'PMID', 'DOI', 'URI', 'shareLink', 'exact', 'rrid', 'public_tags'
]]
nomatch += sorted(
deNone(anno.pmid, anno.doi, anno.uri, anno.shareLink, anno.exact,
anno.rrid, ','.join(
[t for t in anno.public_tags if 'RRID:' not in t]))
for pmid, papers in pmids2.items() for rrids in papers.values()
for annos in rrids.values() for anno in annos
if anno.exact and anno.rrid and anno.exact not in anno.rrid)
with open(f'{ssj}rrids-nomatch-{now}.csv', 'wt') as f:
csv.writer(f, lineterminator='\n').writerows(nomatch)
if args['--debug']:
embed()