What Is bioRxiv and How to Use It (and medRxiv)

bioRxiv (pronounced "bio-archive") is a free server for preprints in the life sciences. A preprint is a research paper that the authors post online before, or instead of, publishing it in a journal, so anyone can read it straight away without a subscription. medRxiv is its sister server for medicine and health research. Between them they hold hundreds of thousands of papers, all free to read, and for recent work in biology and medicine they are often the only free copy there is.

Who runs them

bioRxiv was founded in 2013 at Cold Spring Harbor Laboratory, a research institute in New York. medRxiv followed in 2019 as a joint project of Cold Spring Harbor Laboratory, Yale University and the medical publisher BMJ. In March 2025 both servers moved to a new independent non-profit called openRxiv, set up to run them with a broader base of funding. Nothing changed for readers: the websites, addresses and policies stayed the same.

Neither server charges authors to post or readers to download. They are funded by grants and institutional support, not subscriptions.

What they contain and what they do not

Papers on bioRxiv cover the whole of biology, from neuroscience and genomics to ecology and bioinformatics. medRxiv covers clinical research, epidemiology, public health and related fields. Both accept full research papers, including the figures and supplementary data, but not reviews of other people's work, case reports or opinion pieces.

The important thing to understand is that preprints have not been peer reviewed. Each server screens new submissions before posting, which takes a day or two: staff check that the paper is genuine research in scope, is not plagiarised and does not pose an obvious risk, such as a medical claim that could cause harm if acted on. medRxiv screens more strictly because of that risk. Screening is not a review of whether the science is right. For what peer review does and does not guarantee, see our guide to telling whether an article is peer reviewed.

How to search bioRxiv and medRxiv

Each site has a search box on its home page that takes a title, author name or keywords, with filters for subject area and date. The results pages show the abstract, the date posted and a link to the full text and PDF. There is no account to create.

If you know the paper you want, searching its title in Google Scholar is often faster, because Scholar indexes both servers and shows the preprint next to any published version. Our finder also lists bioRxiv and medRxiv copies when it looks up a paper by DOI, link or title, since the open-access indexes track preprints alongside repository copies.

Versions and DOIs

Every preprint gets a DOI when it is posted. Preprints posted before December 2025 have DOIs starting 10.1101, and newer ones start 10.64898; both keep working. Our guide explains what a DOI is and where to find it.

Authors can upload revised versions, and this is where readers need care. One DOI covers all versions of a paper and always points to the latest one. Older versions remain available at their own addresses, which add "v1", "v2" and so on to the end of the link. If a colleague cites something you cannot find in the paper, check whether they were reading an earlier version.

How to tell whether a preprint was later published

Most preprints are eventually published in a journal, often with changes made during peer review. When bioRxiv or medRxiv matches a preprint to a published article, it adds a notice near the top of the preprint page, "Now published in", with a link to the journal version. The match is automatic and can lag by weeks, so if you see no notice, search the title in Google Scholar or paste the preprint's DOI into our finder, which shows the published version and any free copy of it. Our guide to preprints and published papers explains what may differ between the two.

How to cite a preprint

Cite the preprint, not the published paper, if the preprint is what you read, and say that it is a preprint. A typical citation gives the authors, the year, the title, the server name, the DOI and, ideally, the version:

• Smith J, Lee A. 2025. Title of the paper. bioRxiv. doi:10.1101/2025.01.01.123456v2

Most citation styles now have a rule for preprints; see how to cite a research paper for the main styles. If the paper has since been published, cite the published version instead and check that the finding you are citing survived peer review.

A note of caution on medical preprints

During the COVID-19 pandemic, medRxiv preprints were widely reported in the news as though they were settled findings. Some were later corrected or withdrawn. A preprint is a useful early look at research, and most hold up, but a clinical claim in a preprint has not had the scrutiny a journal article gets. Treat it as provisional, look for the published version, and do not change medical decisions on the strength of a preprint alone.

bioRxiv compared with arXiv

arXiv is the older preprint server for physics, mathematics and computer science, and bioRxiv was modelled on it. They are separate sites with separate searches, so a computational biology paper might be on either, or both. See what arXiv is and how to use it for the other side of the family.