The Awesome Power of Preprint in Genetics
Aug 26, 2013
My PI, Mike Eisen, is famous for his outspoken views on Open Access—including his willingness to put his money where his mouth is: he co-founded PLoS and the lab only publishes in OA journals. In addition to believing that the current system doesn't let enough people read the literature, Mike has long felt that the rate of science publishing is too slow, meaning that interesting results don't appear in print until months, or sometimes years, after the work has been done. This hinders the collaborative (and, let's face it competitive) nature of science, since people who don't go to the exact same conferences are less likely to be exposed to your work. Finally, the fact that your paper is normally seen only by a handful of arbitrarily selected reviewers means that the amount of feedback on your own work is potentially limited as well, especially in the crucial periods where it's still reasonable to adjust your paper to responses.
As we were writing up the results of the first round of experiments in this paper, Mike approached me with the idea of posting the preprint on the arXiv, a preprint server originally for physics research that also hosts quantitative biology papers. I've been a co-author on papers from other labs that took over a year from initial submission to final acceptance (including a resubmission that resets the clock on when it appears to have been "originally submitted"), so I was receptive to the idea that putting up a preprint would be worthwhile. A handful of journals consider posting a preprint as "prior publication", but that list is small and we ultimately decided that any journal that didn't allow posting a preprint was ipso facto not a journal we as a lab wanted to submit to.
The process for getting the paper onto the arXiv was fairly easy, though by no means perfectly painless. I had written my paper in the typesetting program LaTeX so I could use its related citation manager (much easier in my experience than wrestling with Endnote). After that, I just filled in a few metadata fields (authors, abstract, etc), hit submit, and waited until it went live at the end of the current business day. There were a few hiccups with setting it up for PDF output that, if I had either read the instructions more carefully or simply submitted a PDF to begin with wouldn't have been an issue, but overall it wasn't a deal breaker.
Once the paper was live, the hard work began. It's well and good to put a preprint online, but if nobody sees it, then it helps neither them nor you. Thus, we began a minor social media campaign to get the word out. Between posts on Mike's blog and Haldane's Sieve (a blog covering preprints in genetics), Facebook status updates, tweets, and emails to select friends and collaborators, we hit just about every channel that was available. I even mentioned the preprint status in a talk I gave at the Drosophila Conference a month later. While not everyone has a gigantic social media footprint, there's always something you can do to help draw some attention.
Our efforts to reach out seem to have paid off. In the end, we received comments from about 10 scientists, in every career stage from grad students to tenured faculty. The commenters hit just about every end of the self-identification spectrum as well, including pseudonyms, initials, first names only, and full names. Some commented on Mike's blog, some by email, and some in person, and of those who were named, only one or two were people I would have felt comfortable imposing on their time to provide comments on a draft if we weren't posting a preprint.
It's often hard to evaluate whether going the extra mile actually makes a difference, but in this case I got pretty immediate feedback. Literally 48 hours after the paper went live, I got an email from a post-doc at another university who had heard Mike talk about my work and wanted suggestions for dealing with sections prior to sequencing. In addition to answering her specific questions, I was able to point her to the preprint for all the gory details.
All the feedback we got was helpful, either in pointing out controls that needed to be done or simply language that needed to be clarified. When we ultimately submitted to the journal, we told the "official" peer reviewers about our open peer review comments, and the paper sailed through review. I have no doubt this was helped by their seeing my defense of potential shortcomings in the comments, as well as the changes in response to those criticisms.
The other key advantage of posting a preprint was establishing priority. Ultimately, the technique for sequencing really small samples isn't too much of a modification of existing protocols: you add in a carrier RNA that gets sequenced as well, then sort out the sequencing reads at the end. Even granting that the number of labs that could want to scoop us on the particular system was relatively small, Mike and I had both talked about it at enough conferences and seminars that it remained a distinct possibility. Getting it up on the arXiv as quickly as possible would let any potential future employers (I don't want to stay in grad school forever!) know that my paper represented years of work, not a quickly thrown-together copy to apply a new technique to my own pet system.
It's been said that if your ideas are any good, you'll have to ram them down people's throats. Having been through a publishing process with preprints, they seem like such a good idea—especially with straightforward systems that speed your impact, improve your science, and protect your hard work—that people's resistance to good ideas is the only reason I can think of that they aren't the norm in biology.
Copyright: © 2013 Peter Combs. The above content is licensed under the Creative Commons Attribution License (CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
If you find this essay offensive or in violation of your rights, please email to [email protected]