Read our COVID-19 research and news.

New Preprint Server Aims to Be Biologists’ Answer to Physicists’ arXiv

New Preprint Server Aims to Be Biologists' Answer to Physicists' arXiv

A well-known research lab is throwing its weight behind an idea that some biologists say is ripe for their field: a free website that will post raw manuscripts online before they’re submitted to a journal.

BioRxiv, launched yesterday by the nonprofit Cold Spring Harbor Laboratory (CSHL), aims to be biologists' version of arXiv, the popular preprint server where physicists have shared their draft manuscripts for more than 20 years. The goal is to speed the dissemination of research and give scientists a way to get feedback on their papers before they are formally peer-reviewed, says John Inglis, CSHL Press executive director. "There is a growing desire in the community for this kind of service,” Inglis says.

It will be free to submit a paper or to read it in bioRxiv, Inglis says. CSHL is paying the costs of the service (he declines to specify them) but hopes that, like arXiv, it will ultimately attract contributions. Although anybody can submit a paper, not everything will be posted: A group of more than 40 "affiliate" scientists have agreed to screen submissions to "assure us that this is real science," Inglis says. "We certainly don't want the enterprise to be sunk by publishing a load of crap."

Another limitation is that bioRxiv is for life sciences, not medicine, so it will not publish clinical trials or other research that is "medically relevant," Inglis says. Human genetic data could be posted, however.

The site debuts with a handful of papers and a few unusual features. For example, contributors will not only tag their paper with the scientific field, but must also mark it as “New Results,” “Confirmatory Results,” or “Contradictory Results,” depending on whether it is an advance or confirms or contradicts previous experiments. Researchers can post revised versions of the paper and add links to the published paper.

BioRxiv is not a new idea. Fourteen years ago, then-National Institutes of Health (NIH) Director Harold Varmus proposed a preprint server for biology papers. Critics shot it down for numerous reasons, including discomfort with the government hosting such a repository. (The idea, however, led to PubMed Central, NIH’s archive of full-text published manuscripts.)

Preprint servers aimed at biology did eventually appear, including Nature Precedings. It posted thousands of manuscripts over 5 years but closed last year because it became "unsustainable," according to the journal. The Public Library of Science made an attempt at preprints 4 years ago with PLOS Currents: Influenza, but it apparently morphed into a peer-reviewed journal. More recent efforts include PeerJ PrePrints and F1000 Research; the latter site organizes open peer review of the paper and charges a fee.

Skeptics have pointed out that the culture of biology is different from physics—biologists are often reluctant to share their results until they have been validated through peer review. Another concern is that some journals may reject manuscripts that have already been shared publicly. But it depends on the journal. University of California, Los Angeles, geneticist Leonid Kruglyak, an adviser to bioRxiv, keeps a list of journal policies on Wikipedia. (Nature allows preprint posting; Science does in some cases; most Cell journals do not.)

BioRxiv’s 15-member advisory board includes some prominent names, including arXiv creator Paul Ginsparg and Anurag Acharya, co-founder of Google Scholar. Kruglyak says the site has a good shot at success because biology preprints are catching on. In the last couple of years, many quantitative biologists have been sending preprints to arXiv, Kruglyak notes. But that repository "is not designed to appeal to biologists more broadly" and couldn't handle all their papers, he says. BioRxiv’s commenting features will also set it apart from arXiv.

Inglis is a bit more cautious: "We certainly realize that cultural shifts may be required,” he says. If scientists aren't interested, he adds, "we will have learned something."