bioRxiv: the preprint server for biology

Since its 2013 launch, the non-profit preprint server bioRxiv has revolutionized biological research dissemination by enabling immediate sharing of manuscripts across diverse fields, fostering community feedback, and accelerating scientific progress through rapid access and evolving support from journals and funders.

The Gist

Imagine the world of scientific research as a massive, bustling library. For decades, the only way to get a new book (a research paper) onto the shelves was to go through a very strict, slow, and expensive librarian (a scientific journal). This librarian would read the book, argue with the author, make them rewrite chapters, and then, if the book passed, print it and put it on the shelf. This process could take months or even years.

bioRxiv is like a "Pop-Up Book Fair" that scientists set up in their own backyard. Instead of waiting for the strict librarian to approve their books, scientists can put their drafts on a table immediately. Anyone with an internet connection can walk up, read the draft, and leave a sticky note with feedback.

Here is a simple breakdown of how this "Pop-Up Book Fair" works, based on the paper:

1. The Problem: The Waiting Game

In the old days, scientists had to wait for the "Official Librarian" (the journal) to stamp their work before they could share it. This was like baking a cake and not being allowed to tell anyone you made it until a food critic had tasted it, critiqued the frosting, and given it a star rating.

• The Issue: By the time the cake was finally served, the party was over. Scientists, especially young ones trying to get jobs or funding, needed to show they were baking cakes now, not next year.

2. The Solution: The "Preprint" Server

Launched in 2013 by a group called Cold Spring Harbor Laboratory, bioRxiv is a digital shelf where scientists can post their "draft cakes" immediately.

• No Gatekeepers: You don't need the librarian's permission to put your book on the shelf.
• Speed: It takes about 24 to 48 hours to get your paper up there.
• The Catch: The paper is labeled as a "Preprint." It's a real scientific paper, but it hasn't been officially "certified" by a journal yet. It's like saying, "Here is my recipe; I think it's good, but you should taste it yourself before deciding if it's a masterpiece."

3. How It Works: The Safety Net

You might wonder, "What if someone posts nonsense or dangerous recipes?"
bioRxiv has a Safety Team (screeners). They aren't the "taste-testers" (peer reviewers) who decide if the science is perfect. Instead, they are like bouncers at the door. They check:

• Is this actually about biology?
• Is it just a copy-paste job (plagiarism)?
• Is it a recipe for a sandwich instead of a science paper?
  If it passes the bouncer, it goes up. The readers are then responsible for deciding if the science is good.

4. The Feedback Loop: Sticky Notes and Chats

One of the best parts of the Pop-Up Fair is the feedback.

• The Old Way: You sent your paper to a journal, waited months, and got a letter saying "Fix this."
• The bioRxiv Way: You post your paper, and within days, other scientists might email you, tweet about it, or leave comments.
• The Result: A survey in the paper showed that 82% of authors posted their work to get more people to know about it, and 38% wanted feedback. Most authors (89%) said they had zero bad things happen to them by posting early. In fact, it often helped them get jobs, grants, or new friends in their field.

5. The Pandemic Superpower

When the COVID-19 pandemic hit, bioRxiv became a superhero.

• The Situation: The world needed answers about the virus yesterday.
• The Action: Scientists posted their findings on bioRxiv immediately. They didn't wait for the slow journal process.
• The Impact: This allowed doctors and researchers worldwide to see new discoveries in real-time, helping to fight the virus much faster than ever before.

6. The "Handoff" to the Official Library

Does posting on bioRxiv mean you can't publish in a real journal later? No.
Think of bioRxiv as a "trailer" for a movie. You can show the trailer to the world, and then later, the movie studio (the journal) can still release the full film.

• The Transfer: bioRxiv has built a "conveyor belt" (called B2J) that lets authors send their papers from the preprint server directly to a journal with one click. They don't have to re-type everything.
• The Stats: About 80% of papers posted on bioRxiv eventually get published in a traditional journal.

7. The Future: A New Kind of Library

The paper suggests that the way we share science is changing forever.

• Decoupling: We are separating the act of sharing (posting on bioRxiv) from the act of certifying (journal publication).
• The Goal: This makes science faster, more open, and more collaborative. It's like moving from a library where you can only read books after they are bound and stamped, to a community garden where everyone can plant seeds, watch them grow, and help each other water them in real-time.

In a nutshell: bioRxiv is the "fast lane" for science. It lets researchers share their work immediately, get help from the community, and prove they are doing great work, all while waiting for the traditional "slow lane" of journal publishing to catch up.

Technical Summary

Based on the provided paper, here is a detailed technical summary of the bioRxiv preprint server, covering the problem, methodology, key contributions, results, and significance.

1. Problem Statement

The traditional scientific publication process creates significant bottlenecks in the dissemination of research. Key issues include:

• Dissemination Delays: The coupling of evaluation (peer review) and dissemination (publication) often delays the release of new research by months or even years.
• Career Misalignment: These delays conflict with the timelines required for early-career researchers to demonstrate productivity for tenure, grants, and appointments.
• High Costs and Access Barriers: Historically, high marginal costs of print distribution favored a model where only peer-reviewed papers were published, restricting access to non-peer-reviewed drafts.
• Lack of Biology-Specific Infrastructure: While the arXiv server successfully decoupled dissemination from evaluation in physics and mathematics (launched 1991), attempts to replicate this in biology failed due to publisher opposition and low biologist interest until recently.

2. Methodology and Technical Architecture

bioRxiv was launched in 2013 by Cold Spring Harbor Laboratory (CSHL) and is now managed by the non-profit openRxiv. The platform was designed with an "industrial-scale" architecture to handle high volumes of submissions and readership.

• Technical Stack:

  • Submission System: Built on a customized BenchPress module (adapted from Highwire Press), allowing authors to upload PDFs or Word/figure files.
  • Display Engine: Uses modified Highwire JCore technology to render content.
  • Integration: Integrated with third-party tools for commenting (Disqus, Hypothesis) and automated analysis (e.g., Sciscore for resource extraction, Sciencecast for AI summaries).
  • Data Persistence: All content is deposited with Portico for long-term archival preservation.
  • Identifiers: Every preprint is assigned a Digital Object Identifier (DOI). Since 2019, the DOI suffix embeds the author approval date to distinguish versions. Each version has a unique URL, but the DOI defaults to the most recent version.

• Screening Process (Non-Peer Review):

  • bioRxiv explicitly does not perform peer review. Instead, it employs a two-stage coarse screening process to filter out non-scientific, pseudoscientific, harmful, or non-biological content.
  • Stage 1: Automated plagiarism checks (Similarity Check) and spam/incompleteness checks by staff.
  • Stage 2: Manual review by a distributed network of bioRxiv Affiliates (experienced scientists) to ensure the content represents bona fide biological research.

• Interoperability and Transfer:

  • B2J (BioRxiv-to-Journal) & J2B (Journal-to-BioRxiv): Uses standard FTP to transfer XML metadata and manuscript files (ZIP archives) between bioRxiv and journal submission systems, reducing author workload.
  • MECA Adaptation: The transfer protocols are adapted to the Manuscript Exchange Common Approach (MECA).
  • Linking: Automated scripts match preprints to published journal articles via PubMed and Crossref, updating DOI records and displaying links to the "Version of Record."

3. Key Contributions

• Decoupling Dissemination and Evaluation: bioRxiv successfully separated the immediate sharing of research from the formal certification by journals, allowing for real-time scientific communication.
• Scalable Infrastructure: Established a robust, high-availability system capable of processing thousands of submissions monthly and serving millions of views, a feat previously unachieved in the life sciences.
• Pandemic Response: Played a critical role during the SARS-CoV-2 pandemic by rapidly disseminating >30,000 papers on immunology, virology, and epidemiology, often serving as the first source of data for the global scientific community.
• Ecosystem Integration: Created a hub that links preprints to peer reviews (via the TRiP project), community discussions, and automated metrics, fostering a more transparent and interactive scientific dialogue.
• Policy Influence: Helped drive a shift in journal and funder policies, with many major funders (e.g., CZI, Gates Foundation, NIH) now mandating or encouraging preprint posting.

4. Results and Metrics

• Growth: As of the time of writing (Feb 2026), bioRxiv has posted over 310,000 manuscripts. Monthly submissions exceeded 4,000 in 2025.
• Usage: The server receives approximately 7 million abstract views and 4 million PDF downloads per month.
• Demographics: Authors from 209 countries have posted, with the USA, UK, Germany, and China being the top contributors. Top institutions include Stanford, Oxford, and Cambridge.
• Subject Evolution: Initially dominated by evolutionary biology, genomics, and computational biology, the server now sees significant growth in neuroscience, cell biology, and developmental biology.
• Publication Rate: Approximately 80% of bioRxiv preprints are eventually published in traditional journals within a 3-year window. The average time from preprint posting to journal publication is 256 days (median 204 days).
• Withdrawals: The withdrawal rate is low (0.17%), comparable to journal retractions (~0.2%). Withdrawn papers remain accessible via history tabs but are clearly marked.
• User Survey Findings (2023/2024):
  • Motivations: Primary drivers include increasing research awareness (82%), staking priority claims (56%), and controlling availability (54%).
  • Feedback: 36% of authors receive feedback via email, and 34% via direct conversation. Only 4% receive on-site comments, though 46% express a desire for it.
  • Negative Consequences: 89% of authors reported no negative consequences from posting preprints. Only 2% felt it hindered publication in specific journals.
  • Timing: 55% post preprints around the time of journal submission, while 30% post weeks to months prior.

5. Significance

• Acceleration of Science: By enabling immediate sharing, bioRxiv accelerates the pace of research, allowing scientists to build upon new findings without waiting for formal publication.
• Cultural Shift: It has normalized the practice of preprinting in the life sciences, moving it from a fringe activity to a standard component of the research workflow.
• Evolution of Peer Review: The existence of preprints is driving experimentation with new peer review models (e.g., transparent review, portable review) and reducing the pressure on journals to be the sole gatekeepers of scientific validity.
• Equity and Openness: It contributes to a more open and equitable system by making research freely available immediately, bypassing paywalls and geographic barriers associated with traditional journals.
• Future Outlook: The platform is evolving to accommodate new data types (code, notebooks) and leveraging AI for screening and summarization, signaling a shift away from the static "paper" format toward dynamic, data-rich research outputs.