def get_response_dict(id_type, ids, pmc_query):
response_dict = {}
num_ids = len(ids)
id_counter = 0
geo_instance = geo.GEO()
for accession in ids:
print accession,
id_counter += 1
reuse_pmcids = get_accession_in_pmc_fulltext(id_type, accession,
pmc_query)
if ((not reuse_pmcids) or ("<ERROR>" in "".join(reuse_pmcids))):
print " Nope"
continue
stripped_accession = geo.get_stripped_accession(accession)
if id_type == "GSE":
gse_accessions = [stripped_accession]
gds_accessions = [
geo.get_stripped_accession(acc)
for acc in geo.get_gds_from_gse("GSE" + stripped_accession)
]
else:
gds_accessions = [stripped_accession]
gse_accessions = [
geo.get_stripped_accession(acc)
for acc in geo.get_gse_from_gds("GDS" + stripped_accession)
]
try:
submit_pmids = geo_instance.pmids("GSE" + gse_accessions[0])
except Exception:
continue
print id_counter, "of", num_ids, ":", stripped_accession, "--", (
submit_pmids), "; ", len(reuse_pmcids)
for reuse_pmcid in reuse_pmcids:
reuse_pmids_for_pmcid = pubmedcentral.pmcids_to_pmids(reuse_pmcid)
dict_key = (id_type + stripped_accession, reuse_pmcid)
this_submit_contributors = flatten_unique([
geo_instance.contributors("GSE" + gse_accession)
for gse_accession in gse_accessions
])
this_submit_authors = get_authors_and_submittors_from_accession(
submit_pmids, this_submit_contributors)
this_reuse_authors = get_authors_and_submittors_from_accession(
reuse_pmids_for_pmcid)
intersect = get_author_intersect_submit_reuse(
this_submit_authors, this_reuse_authors)
submit_affiliation = affiliation.institution(submit_pmids)
release_date = geo_instance.release_date("GSE" + gse_accession)
response_dict[dict_key] = (id_type + stripped_accession,
gse_accessions, gds_accessions,
submit_pmids, reuse_pmcid,
reuse_pmids_for_pmcid,
this_submit_authors, this_reuse_authors,
intersect, submit_affiliation,
release_date)
#print response_dict[dict_key]
return (response_dict)
def get_all_attributes(id_dict):
all_accession_key = [vals[0] for vals in id_dict.values()]
all_submit_pmids = flatten_unique([vals[3] for vals in id_dict.values()])
all_reuse_pmcids = flatten_unique([vals[4] for vals in id_dict.values()])
all_reuse_pmids = flatten_unique([vals[5] for vals in id_dict.values()])
reuse_affiliation = affiliation.institution(all_reuse_pmids)
journal = pubmed.journal(all_reuse_pmids)
year = pubmed.year_published(all_reuse_pmids)
date_published = pubmed.date_published(all_reuse_pmids)
medline_status = pubmed.medline_status(all_reuse_pmids)
is_geo_reuse = geo_reuse.is_geo_reuse(all_reuse_pmids)
reuse_is_oa = pubmed.is_open_access(all_reuse_pmids)
metaanal = pubmed._get_flags_for_pattern(all_reuse_pmids, metaquery)
oa_excerpts = oaexcerpt.get_oa_excerpts(all_reuse_pmcids, "(GSE.\d|GDS.\d|omnibus)", flags = re.IGNORECASE|re.MULTILINE)
biolink_filter = pubmedcentral.get_flags_for_pattern(all_reuse_pmcids, '(geo OR omnibus) AND microarray AND "gene expression" AND accession NOT (databases OR user OR users OR (public AND accessed) OR (downloaded AND published))')
basic_reuse_filter = pubmedcentral.get_flags_for_pattern(all_reuse_pmcids, '"gene expression omnibus" AND (submitted OR deposited)')
creation_filter = pubmedcentral.get_flags_for_pattern(all_reuse_pmcids, '("gene expression" [text] AND "microarray" [text] AND "cell" [text] AND "rna" [text]) AND ("rneasy" [text] OR "trizol" [text] OR "real-time pcr" [text]) NOT ("tissue microarray*" [text] OR "cpg island*" [text])')
has_mesh = {}
for term in meshes:
has_mesh[term] = pubmed.filter_pmids(all_reuse_pmids, term)
mesh_filters = [";".join([term for term in has_mesh if pmid in has_mesh[term]]) for pmid in all_reuse_pmids]
has_word = {}
for word in words:
has_word[word] = pubmedcentral.filter_pmcids(all_reuse_pmcids, word)
word_filters = [";".join([word for word in has_word if pmcid in has_word[word]]) for pmcid in all_reuse_pmcids]
print word_filters
reuse_pmid_dict = defaultdict(tuple, zip(all_reuse_pmids, zip(reuse_affiliation, journal, year, date_published, medline_status, is_geo_reuse, reuse_is_oa, metaanal, mesh_filters)))
reuse_pmcid_dict = defaultdict(tuple, zip(all_reuse_pmcids, zip(biolink_filter, basic_reuse_filter, creation_filter, oa_excerpts, word_filters)))
full_dict = {}
for vals in id_dict.values():
id = vals[0]
reuse_pmcid = vals[4]
reuse_pmid = vals[5][0] if vals[5] else ""
full_dict[id+reuse_pmcid] = vals + ("|",) + reuse_pmcid_dict[reuse_pmcid] + ("|",) + reuse_pmid_dict[reuse_pmid]
print full_dict[id+reuse_pmcid]
return(full_dict)
def get_response_dict(id_type, ids, pmc_query):
response_dict = {}
num_ids = len(ids)
id_counter = 0
geo_instance = geo.GEO()
for accession in ids:
print accession,
id_counter += 1
reuse_pmcids = get_accession_in_pmc_fulltext(id_type, accession, pmc_query)
if ((not reuse_pmcids) or ("<ERROR>" in "".join(reuse_pmcids))):
print " Nope"
continue
stripped_accession = geo.get_stripped_accession(accession)
if id_type=="GSE":
gse_accessions = [stripped_accession]
gds_accessions = [geo.get_stripped_accession(acc) for acc in geo.get_gds_from_gse("GSE"+stripped_accession)]
else:
gds_accessions = [stripped_accession]
gse_accessions = [geo.get_stripped_accession(acc) for acc in geo.get_gse_from_gds("GDS"+stripped_accession)]
try:
submit_pmids = geo_instance.pmids("GSE"+gse_accessions[0])
except Exception:
continue
print id_counter, "of", num_ids, ":", stripped_accession, "--", (submit_pmids), "; ", len(reuse_pmcids)
for reuse_pmcid in reuse_pmcids:
reuse_pmids_for_pmcid = pubmedcentral.pmcids_to_pmids(reuse_pmcid)
dict_key = (id_type+stripped_accession, reuse_pmcid)
this_submit_contributors = flatten_unique([geo_instance.contributors("GSE"+gse_accession) for gse_accession in gse_accessions])
this_submit_authors = get_authors_and_submittors_from_accession(submit_pmids, this_submit_contributors)
this_reuse_authors = get_authors_and_submittors_from_accession(reuse_pmids_for_pmcid)
intersect = get_author_intersect_submit_reuse(this_submit_authors, this_reuse_authors)
submit_affiliation = affiliation.institution(submit_pmids)
release_date = geo_instance.release_date("GSE"+gse_accession)
response_dict[dict_key] = (id_type+stripped_accession, gse_accessions, gds_accessions, submit_pmids, reuse_pmcid, reuse_pmids_for_pmcid, this_submit_authors, this_reuse_authors, intersect, submit_affiliation, release_date)
#print response_dict[dict_key]
return(response_dict)
def get_all_attributes(id_dict):
all_accession_key = [vals[0] for vals in id_dict.values()]
all_submit_pmids = flatten_unique([vals[3] for vals in id_dict.values()])
all_reuse_pmcids = flatten_unique([vals[4] for vals in id_dict.values()])
all_reuse_pmids = flatten_unique([vals[5] for vals in id_dict.values()])
reuse_affiliation = affiliation.institution(all_reuse_pmids)
journal = pubmed.journal(all_reuse_pmids)
year = pubmed.year_published(all_reuse_pmids)
date_published = pubmed.date_published(all_reuse_pmids)
medline_status = pubmed.medline_status(all_reuse_pmids)
is_geo_reuse = geo_reuse.is_geo_reuse(all_reuse_pmids)
reuse_is_oa = pubmed.is_open_access(all_reuse_pmids)
metaanal = pubmed._get_flags_for_pattern(all_reuse_pmids, metaquery)
oa_excerpts = oaexcerpt.get_oa_excerpts(all_reuse_pmcids,
"(GSE.\d|GDS.\d|omnibus)",
flags=re.IGNORECASE | re.MULTILINE)
biolink_filter = pubmedcentral.get_flags_for_pattern(
all_reuse_pmcids,
'(geo OR omnibus) AND microarray AND "gene expression" AND accession NOT (databases OR user OR users OR (public AND accessed) OR (downloaded AND published))'
)
basic_reuse_filter = pubmedcentral.get_flags_for_pattern(
all_reuse_pmcids,
'"gene expression omnibus" AND (submitted OR deposited)')
creation_filter = pubmedcentral.get_flags_for_pattern(
all_reuse_pmcids,
'("gene expression" [text] AND "microarray" [text] AND "cell" [text] AND "rna" [text]) AND ("rneasy" [text] OR "trizol" [text] OR "real-time pcr" [text]) NOT ("tissue microarray*" [text] OR "cpg island*" [text])'
)
has_mesh = {}
for term in meshes:
has_mesh[term] = pubmed.filter_pmids(all_reuse_pmids, term)
mesh_filters = [
";".join([term for term in has_mesh if pmid in has_mesh[term]])
for pmid in all_reuse_pmids
]
has_word = {}
for word in words:
has_word[word] = pubmedcentral.filter_pmcids(all_reuse_pmcids, word)
word_filters = [
";".join([word for word in has_word if pmcid in has_word[word]])
for pmcid in all_reuse_pmcids
]
print word_filters
reuse_pmid_dict = defaultdict(
tuple,
zip(
all_reuse_pmids,
zip(reuse_affiliation, journal, year, date_published,
medline_status, is_geo_reuse, reuse_is_oa, metaanal,
mesh_filters)))
reuse_pmcid_dict = defaultdict(
tuple,
zip(
all_reuse_pmcids,
zip(biolink_filter, basic_reuse_filter, creation_filter,
oa_excerpts, word_filters)))
full_dict = {}
for vals in id_dict.values():
id = vals[0]
reuse_pmcid = vals[4]
reuse_pmid = vals[5][0] if vals[5] else ""
full_dict[id + reuse_pmcid] = vals + (
"|", ) + reuse_pmcid_dict[reuse_pmcid] + (
"|", ) + reuse_pmid_dict[reuse_pmid]
print full_dict[id + reuse_pmcid]
return (full_dict)
