refid2bib

A simple tool get get bibtex entries based on a variety of common academic ids. It's not the only such thing out there (I was largely inspired by doi2bib), but I wanted something that could handle Biorxiv citations in a manner that was parallel with arXiv citations.

How to use:

From a python console/notebook:

There's only one function to use, refid2bib, which will attempt to intuit your request. It can handle DOIs, Arxiv Ids, Pubmed ids, PMC ids, and Biorxiv ids. See test_refid2bib.py for various examples.

For example, a typical doi copied from Pubmed:

from refid2bib import refid2bib
doi = "DOI: 10.1126/science.1260088"
bibtex = refid2bib(doi)
print(bibtex)

will return

@article{Chen_2015,
    doi = {10.1126/science.1260088},
    url = {https://doi.org/10.1126%2Fscience.1260088},
    year = 2015,
    month = {jan},
    publisher = {American Association for the Advancement of Science ({AAAS})},
    volume = {347},
    number = {6221},
    pages = {543--548},
    author = {Chen, F. and Tillberg, P. W. and Boyden, E. S.},
    title = {Expansion microscopy},
    journal = {Science}
}

To specify the reference name that will be used, use the short_name parameter like so

doi = "https://doi.org/10.1126/science.1260088"
custom_name = 'my_custom_name'
bibtex = refid2bib(doi, short_name=custom_name)
print(bibtex)

to return:

@article{my_custom_name,
    doi = {10.1126/science.1260088},
    ...

For author names if you would prefer to use (First Middle Last Suffix) order instead of the default (Last Suffix, First Middle), set the lastname_first parameter to False.

doi = "https://doi.org/10.1126/science.1260088"
bibtex = refid2bib(doi, lastname_first=False)
print(bibtex)

This returns the same thing, but now the author line is:

    author = {F. Chen and P. W. Tillberg and E. S. Boyden},

From the command line:

refid2bib can be run as a function with a reference passed as an argument like so:

python -m refid2bib https://doi.org/10.1101/406314

Note that the custom name and author name order options are not yet implemented in the command line version.

Format Detection

Case does not matter for any reference.

DOI: A doi needs to either start with https://doi.org/ or have a leading doi: (with or without the space). The id is taken to be everything after the slash of the URL or the space/colon. For example,doi: 10.1126/science.1260088 or https://doi.org/10.1126/science.1260088.

Biorxiv: Biorxiv entries are detected if they start with a biorxiv: (with or without the space) or have a DOI that corresponds to Biorxiv (i.e. a CSHL 10.1101 prefix and a purely numeric suffix). This mode adds eprint and eprinttype entries to the bibtex output, similar to Arxiv entries. For example, biorxiv:376830 or https://doi.org/10.1101/376830. These are returned as @online types, not articles.

ArXiv: Arxiv entries are detected if they start with arxiv: (no space) or from the URL directly https://arxiv.org/abs/. For example, arXiv:1801.04381 or https://arxiv.org/abs/1801.04381. These are returned as @online types, not articles.

Pubmed: Either pubmed ids or PMC ids will work, either from pmid: (with or without space) or PMC (no space) respectively. For example, PMID: 25056931 or PMC3711719.