1
#V2.0
2
#by m36-intj: makes bibtex entries from a list of pubmed article pmids, uses pmc article link, if available
3
#fixed the problem with sometimes wrong publication date, uses now biopython instead of pymed to get publication date
4
#same with title, uses now biopython for titles
5
#added unicode to latex translation of author names
6
#uses now regex special rules for title conversion to latex format
7
 
8
#input list file format is
9
#%comment
10
#12345678
11
#23415251
12
#....
13
 
14
from pymed import PubMed
15
pubmed = PubMed()
16
from Bio import Entrez
17
from Bio import Medline
18
from pylatexenc.latexencode import unicode_to_latex
19
from pylatexenc.latexencode import UnicodeToLatexEncoder, UnicodeToLatexConversionRule, RULE_REGEX
20
import re
21
 
22
Entrez.email = "[email protected]"  # Required by NCBI
23
 
24
def test():
25
    output_file = open('/storage/emulated/0/Download/python/bibtex.txt', "w")
26
    pmids= open('/storage/emulated/0/Download/python/pmids', "r")
27
    
28
    for pmid in pmids:
29
    	if pmid.startswith("%"):
30
    		continue
31
    	pmid = pmid.strip()
32
    	results = pubmed.query(f"{pmid}[PMID]", max_results=1)
33
    	article = next(results, None)
34
    	if article:
35
    	   print(f"PMID: {pmid}")
36
    	   title = fetch_title(pmid)
37
    	   title = special_convert(title)
38
    	   print(f"Title: {title}")
39
    	   raw_id = str(article.pubmed_id)
40
    	   clean_pmid = raw_id.split()[0]  # Takes first entry if multiple exist
41
    	   
42
    	   first_author = article.authors[0]  # Get first author dict
43
    	   first_author_name = f"{first_author.get('firstname', '')} {first_author.get('lastname', '')}".strip()
44
    	   #translate to latex syntax
45
    	   first_author_name = unicode_to_latex(first_author_name)
46
    	   print(f"First Author: {first_author_name}")
47
    	   
48
    	   url = get_pmc(clean_pmid)
49
    	   print(f"URL: {url}")
50
    	   # Extract year 
51
    	   handle = Entrez.esummary(db="pubmed", id=clean_pmid)
52
    	   record = Entrez.read(handle)
53
    	   handle.close()
54
    	   year=record[0]["PubDate"].split()[0]
55
    	   print(f"Publication year: {year}")
56
    
57
    	   # get ISO abbreviation for journal name
58
    	   abbrev = get_journal_abbreviation(clean_pmid)  # Replace with your PMID
59
    	   print(f"ISO Abbreviation: {abbrev}")  # e.g., "Nat. Biotechnol."
60
    	   #make a bibtex entry
61
    	   bibtex_entry(output_file,first_author_name, url, title, abbrev, year)
62
    	   print("")
63
    	else:
64
    		print(f"\nPMID {pmid} not found.")
65
    
66
# Example:
67
def get_pmc(pmid):
68
        try:
69
        	handle = Entrez.elink(dbfrom="pubmed", db="pmc", id=pmid)
70
        	record = Entrez.read(handle)
71
        	if record and "LinkSetDb" in record[0] and record[0]["LinkSetDb"]:
72
        	   	pmcid = record[0]["LinkSetDb"][0]["Link"][0]["Id"]
73
        	   	return f"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC{pmcid}/"
74
        except Exception as e:
75
        	print(f"Error fetching PMCID for PMID {pmid}: {str(e)}")
76
        return  f"https://pubmed.ncbi.nlm.nih.gov/{pmid}"
77
    
78
 
79
def get_journal_abbreviation(pmid):
80
    
81
    try:
82
        # Step 1: Fetch record in Medline format (more reliable than XML)
83
        handle = Entrez.efetch(db="pubmed", id=pmid, rettype="medline", retmode="text")
84
        record = Medline.read(handle)
85
        
86
        # Step 2: Check all possible journal name fields
87
        journal_fields = ['TA', 'JT', 'SO']
88
        for field in journal_fields:
89
            if field in record:
90
                return record[field]
91
        
92
        # Step 3: Ultimate fallback
93
        return record.get('SO', 'Journal name unavailable')
94
    
95
    except Exception as e:
96
        print(f"Error processing PMID {pmid}: {str(e)}")
97
        return "Journal name unavailable"
98
 
99
def fetch_title(pmid):
100
    """Fetch article title for a given PMID."""
101
    try:
102
        # Fetch XML data for the PMID
103
        handle = Entrez.efetch(db="pubmed", id=pmid, retmode="xml")
104
        record = Entrez.read(handle)
105
        
106
        # Extract title from the nested XML structure
107
        if "PubmedArticle" in record:
108
            article = record["PubmedArticle"][0]
109
            title = article["MedlineCitation"]["Article"]["ArticleTitle"]
110
            return title
111
    except Exception as e:
112
        print(f"Error fetching title for PMID {pmid}: {e}")
113
    return None
114
 
115
 #this converts strings to LaTex format, there might be special rules which are needed e.g. <i>text</i> -> \textit{text}
116
def special_convert(string):
117
	u = UnicodeToLatexEncoder(
118
	conversion_rules=[
119
	UnicodeToLatexConversionRule(rule_type=RULE_REGEX, rule=[
120
	(re.compile(r'<i>'), r'\\textit{'),
121
	(re.compile(r'</i>'), r'}'),
122
	]),
123
	'defaults'
124
	])
125
	return u.unicode_to_latex(string)
126
	
127
#@Article{hallmarks_aging,
128
# Author         = "Carlos Lopez-Otin and others",
129
# Title          = "\href{https://pubmed.ncbi.nlm.nih.gov/36599349/}{Hallmarks of aging: An expanding universe}",
130
# Journal        = "Cell Metabolism",
131
# Year           = 2023,
132
#}
133
def bibtex_entry(text_file, author, url, title, journal, year):
134
		print(f"@Article {{key,",file=text_file)
135
		print(f"Author = \"{author} and others\",",file=text_file)
136
		print(f"Title = \"\\href{{{url}}}{{{title}}}\",",file=text_file)
137
		print(f"Journal = \"{journal}.\",",file=text_file)
138
		print(f"Year = {year},\n}}\n",file=text_file)
139
		return
140
 
141
test()