In March, a group of physicists published an incredible scientific result. A result that has fundamentally changed our understanding of nature and enables scientists to study the very origins of our universe. I am talking here about the discovery  of direct evidence of primordial gravitational waves, literally ripples in the fabric of space-time, that are a key prediction of Einstein’s theory of general relativity. It was announced to tremendous fanfare on March 17, being described as “a sensational breakthrough”, “one of the greatest discoveries in the history of modern science”, and yet there is an interesting caveat. When it was announced, it had not yet been peer reviewed.
Irrespective of how exciting a result may be, to publicly announce your findings ahead of proper peer review is far from the norm in biology. Yet this is common practice in physics and mathematics circles. The key difference between these communities is the acceptance amongst physics researchers of online, pre-publication resources, the most famous of which can be found at arxiv.org
The ArXiv (pronounced “archive”, as in the greek letter chi) is an online repository for electronic preprints of scientific papers. It was originally set up in 1991 by Paul Ginsparg, and has since become a huge resource, in some fields being the de facto means of releasing results. More than half a million preprints are accessible, free of charge, and that number grows by more than 6000 papers per month.
The pressure to see results published is huge in science, yet it is not uncommon to spend months or even years with a paper ready to be published as it wades through the peer review process. Furthermore it is common to have to resubmit an article to various scientific journals (requiring sometimes extensive reformatting at each step) before that paper sees the light of day. In the mean time, your competitors have the opportunity to slip in ahead of you. Pre-publication not only allows your research to get out into the wider community much earlier, it also allows members of competitive fields to clearly recognise whose work takes precedence instead of being held at the mercy of editors and the review process.
The peer review system is a fundamental strength of science. It is through this procedure, as frustrating as it may be, that scientists and more importantly the public know that research is, to the best of our ability, trustworthy. To some the idea of pre-publication invites abuse, and risks undermining the trust that is central to the scientific method. How does a layperson recognise the difference between a pre-published manuscript on the ArXiv and the peer reviewed publication in a high end journal? Particularly in the case of headline grabbing science such as the gravitational wave story.
An interesting benefit of the ArXiv is that other scientists are able to provide feedback on the work before it is published. Rather than relying on the comments of 3 or so reviewers, the entire community has the opportunity to discuss the work. The more headline grabbing the paper, the greater the interest will be, and so the more discussion would be generated. Furthermore, other members of the field can immediately work on verifying results before it is published in a journal. It is feasible to imagine that situations such as the recent induction of pluripotency by acid stress, published in Nature on January 30  and almost immediately embroiled in controversy, could be avoided.
It is exciting to see that the first steps are being made towards acceptance of pre-publication in biology. Journals such as F1000Research and PeerJ have been in existence for a little while now. F1000Research and PeerJ are both journals that make articles available immediately upon receipt and then submit them to a standard peer review system. Where these systems are limited though, is that you are committing to publish your article with them. The current landscape is too heavily influenced by impact factor metrics (a problem in of itself) for pre-publication ideals to make headway. If you have ground breaking results, it is more valuable for your career to go to Nature or Science despite the wait for publication.
Where the ArXiv succeeds is that it is not tied to a particular journal. Papers submitted to the ArXiv are still recognised and accepted by major journals, including Nature and Science. Researchers submitting manuscripts to the ArXiv do not risk losing out on high impact journals. Sadly, the ArXiv has no plans to extend its remit to the biological sciences (beyond the mathematically orientated fields such as bioinformatics and systems biology). That is where the BioRxiv steps in.
The BioRxiv (pronounced “bio-archive”) opened in December of 2013 as the biological equivalent of the ArXiv. While it is run independent of a particular journal (it was set up by Cold Spring Harbour), it has taken the ArXiv as a model, unlike journals such as F1000Research and PeerJ. Indeed, Paul Ginsparg is listed as being on the advisory board, which is a powerful endorsement if nothing else. It is already gaining support amongst some members of the community (see this article for example), but there is still a long way to go.
Today it is widely accepted amongst scientists that Open Access is the way forward, even if battles are still being raged. It is likely that the next revolution in science will focus on the peer review system. There is no magic cure available, and no perfect utopia. We as a community are, however, on the first steps of that journey and it is up to us where we want that journey to end. Pre-publication will almost inevitably be a part of that destination, but what form that will take is unclear. The BioRxiv as modelled on the tremendously successful ArXiv has an amazing opportunity to be there, if we choose to support it.
For more information about the BioRxiv, check it out here
 arXiv:1403.3985 [astro-ph.CO]
 Obokata, H. et al. Nature 505, 676–680 (2014).
By Matthew Evans, a second year PhD student in the lab of Richard Morris