This article appeared in the Spring 2013 issue of Current Exchange Magazine.
It’s customary in my lab to check the top science journals weekly to see what articles have been posted to the advanced online publication sections. It isn’t uncommon for me to arrive at lab in the morning and be greeted with,
“Have you read so-and-so’s new paper?”
“No. When was it published?”
Point being, scientists are eager to access new scientific data as soon as it’s made available. The latest findings motivate new research questions and fill in gaps in knowledge, helping scientists to design experiments and interpret the results of work in progress. While advance online publication makes articles available as soon as the peer review process is complete, that process can still take many months. Imagine accessing an early version of a manuscript at the same time, or even before, it’s been received by journal editors.
bioRxiv allows research scientists to upload and read (for free) manuscripts prior to peer review and publication, thus disseminating results freely and rapidly. Once a manuscript is posted to bioRxiv, it receives a citable web address and is discoverable in web searches. Scientists can then comment on the merits and weaknesses of the manuscript on bioRxiv, potentially influencing how the final version appears. This model is based in part on Cornell University’s arXiv service, which has been a mainstay of the math and physics research communities for the past two decades.
I talked to Dr. John Inglis, Executive Director and Publisher of CSHL Press about his plans for bioRxiv.
The first conversation about the role of a preprint server in biology took place between me and Paul Ginsparg (the founder of arXiv) about 10 years ago. My conclusion at that stage was that it would not work in biology as it was at that time. More recently, we’ve been tracking the increase in amount of biology that’s been posted to arXiv.
You can see a dramatic increase in the volume of material posted to the quantitative biology section, which is the only section they have that addresses biology. ArXiv was aware of this trend but had no problem with our attempting to build a complementary site that had features and functions that were more familiar to people in mainstream biology. In fact, Paul Ginsparg became the first member of our advisory board. While we have no formal association with arXiv we are very similar in terms of our goals and aspirations, and hope to be talking to them as we get further along.
A manuscript can go through a very lengthy process of review at a given journal and then the answer can be “no thanks.” In the meantime, a draft of that manuscript can be on the bioRxiv server for the benefit of the broader community who now have access to the information it contains. It has to be taken skeptically because it has not been peer-reviewed, but its contents may be valuable and should be critically evaluated. Eventually one hopes that the manuscript will be adapted for the purposes of a journal, and it will find its way into the formal literature. We keep saying at every opportunity that bioRxiv is not a journal, with the quality assurance and imprimatur that journals traditionally provide. It’s a tool for the rapid distribution of research results and feedback on work in progress.
Through its conferences, courses, and publications, Cold Spring Harbor Laboratory has provided a service of communication to the scientific community for many decades, going back to 1933 and the first Symposium in Quantitative Biology. We see bioRxiv as an extension of that: a new tool for professional communication amongst scientists, one that operates in the digital world and involves critical evaluation, skepticism, attack and defense – all the sorts of interactions that are on display at meetings here on the campus.
We have two levels of scrutiny. One of them being a purely clerical one to ensure that what has been submitted is not obscene or spam. Then we have a growing community of bioRxiv Affiliates, working scientists who have agreed to just look at submitted manuscripts without evaluating their quality and identify them as science rather than non-science or pseudoscience. That’s doable at this early stage when the number of submitted manuscripts is still manageable.
We have two ways of tagging the material that’s deposited – one is to put it in a subject category, for example, “genomics”. The other one is more interesting – we’re asking authors to flag as “new results”, “confirmatory results”, and “contradictory results” to give the reader a little bit of extra insight into what the significance of the manuscript is. That’s another part of the service we consider this server can provide. It’s difficult to get confirmatory and contradictory results published formally in journals, so this is a way of sharing that information.
In the beginning it’s being supported by the Laboratory. You’ll see when you look at the site that it’s pretty basic; it’s not sophisticated, not slick. We haven’t spent a fortune on graphic design. It’s essentially a bunch of manuscripts formatted as PDFs, with supplementary data where appropriate but no bells, no whistles. You take it as you find it. I have held back from doing anything about outside financial support until we get up and running. Soon I’ll start a round of conversations with organizations that might help us with the running costs.
We’re not aligning ourselves with any movement at all. What we’re doing is responding to growing interest and desire within the scientific community for different forms of communication that are more open, more immediate, and in many ways more unfiltered than the conventional journal system is. As the publisher of two of the top three genetics journals, CSHL Press is very invested in the value that journals bring to the scientific process and we’re very proud of the peer review that takes place at our journals.
There is a tendency at the moment to dismiss peer review as obstructive or harmful in some way, but I’m not willing to take that position. Is it perfect? No, but I see a lot of people putting an enormous amount of effort into that process, and I think that’s very valuable.
I hope bioRxiv will become part of the ecosystem of scientific communication and be judged and valued for what it is, not for what it isn’t. I think journals will continue to have an important role in the filtration and validation processes they have always had. I don’t see that bioRxiv will necessarily undermine that role, but I do think it can assist in a variety of other ways with the rapid dissemination of results, communication of work in progress, and facilitation of community feedback, and doing all of that in an open way. That’s a little different than the concept of open access that is changing so much of how journals operate.
We want to communicate with the scientific community in its widest sense. What’s going to capture attention obviously requires scale, and scale on bioRxiv requires a culture shift within biomedical science. What would upset me most is that there is a knee-jerk negative reaction to bioRxiv that doesn’t allow it the chance to reach its full potential. Obviously we want this to be a success and will work hard to make it so, but ultimately it’s the research community that will determine that.
At this stage we don’t know all the ways that bioRxiv will function. It’s a new idea for most people in biology, and I think it’s going to be fascinating to see how the community deals with it.
bioRxiv launched on November 11, 2013 and is currently accepting submissions. You can learn more at biorxiv.org, and see a list of academic journals by preprint policy curated by bioRxiv advisory board member Dr. Leonid Krugylak.