def generate_all_article_info(csv_dir="1. excel files"):
'''
1. Parse all .csv files in the directory specified by csv_dir, and,
2. Generate corresponding article information files that include
PMIDs and titles (citation info) for articles annotated in csv
files.
'''
failures = []
for csv_f in [f for f in os.listdir(csv_dir) if f.endswith(".xls")]:
cur_title = get_article_title(os.path.join(csv_dir, csv_f))
#get pubmid from title
pmid = pubmedpy.get_pmid_from_title(cur_title)
if pmid:
print "matched %s for %s." % (pmid, csv_f)
citation_info = list(pubmedpy.fetch_articles([pmid]))[0]
title, mesh, authors, abstract, affiliation, journal, volume, article = get_fields_from_article(
citation_info)
with open(
os.path.join(csv_dir,
csv_f).replace(".xls", ".article_info.txt"),
'wb') as out_f:
csv_writer = csv.writer(out_f)
csv_writer.writerow([
pmid, title, mesh, authors, abstract, affiliation, journal,
volume
])
else:
#print "!!! failure !!! could not find PubMed entry for %s" % cur_title
failures.append(cur_title)
print "ok! %s failures (see below, if any). \n %s" % (len(failures),
"\n".join(failures))
def generate_all_article_info(csv_dir="1. excel files"):
'''
1. Parse all .csv files in the directory specified by csv_dir, and,
2. Generate corresponding article information files that include
PMIDs and titles (citation info) for articles annotated in csv
files.
'''
failures = []
for csv_f in [f for f in os.listdir(csv_dir) if f.endswith(".xls")]:
cur_title = get_article_title(os.path.join(csv_dir, csv_f))
#get pubmid from title
pmid = pubmedpy.get_pmid_from_title(cur_title)
if pmid:
print "matched %s for %s." % (pmid, csv_f)
citation_info = list(pubmedpy.fetch_articles([pmid]))[0]
title, mesh, authors, abstract, affiliation, journal, volume, article = get_fields_from_article(citation_info)
with open(os.path.join(csv_dir,csv_f).replace(".xls", ".article_info.txt"), 'wb') as out_f:
csv_writer = csv.writer(out_f)
csv_writer.writerow([pmid, title, mesh, authors, abstract, affiliation, journal, volume])
else:
#print "!!! failure !!! could not find PubMed entry for %s" % cur_title
failures.append(cur_title)
print "ok! %s failures (see below, if any). \n %s" % (len(failures), "\n".join(failures))
with open(FILE_PATH, 'rU') as f:
reader = csv.DictReader(f, dialect='excel', delimiter='\t')
with open(OUT_PATH, 'w') as csvfile:
fieldnames = [hPmid, hCitation, hTitle, hPrediction, hHard]
writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
for row in reader:
output_row = {}
citation_id = row[hCitation]
title = row[hTitle]
predicted = row[hPrediction]
hard = row[hHard]
try:
pmid = pubmedpy.get_pmid_from_title(title)
pmid = int(pmid)
except Exception as e:
pmid = '-- Cannot find --'
output_row = {
hPmid: pmid,
hCitation: citation_id,
hTitle: title,
hPrediction: predicted,
hHard: hard
}
writer.writerow(output_row)
###############################################################################################
with open(FILE_PATH, 'rU') as f:
reader = csv.DictReader(f, dialect='excel', delimiter='\t')
with open(OUT_PATH, 'w') as csvfile:
fieldnames = [hPmid, hCitation, hTitle, hPrediction, hHard]
writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
for row in reader:
output_row = {}
citation_id = row[hCitation]
title = row[hTitle]
predicted = row[hPrediction]
hard = row[hHard]
try:
pmid = pubmedpy.get_pmid_from_title(title)
pmid = int(pmid)
except Exception as e:
pmid = '-- Cannot find --'
output_row = { hPmid: pmid, hCitation: citation_id,
hTitle: title, hPrediction: predicted,
hHard: hard }
writer.writerow(output_row)
###############################################################################################
#### This is fixing Emily's project. The refman ids weren't saved when she imported the project
#PROJECT_ID = 219
#FILE_PATH = './Abstraktr_Update_Lit_Review_11.12.13.txt'
#citations_q = Session.query(model.Citation)