Bio Archive - Thoughts?

[Kron3007]

It is not common in my field yet, but I was discussing bio archive (BioRxiv) with a researcher in a related field that uses it a lot and am thinking about starting.  I just wanted to get peoples' thoughts on it.

For those who are not familiar, I will give you my potentially flawed version.  It is a pre-print site hosted by Cold Spring Harbour Laboratory (important to me as it is not a for profit organization) where you can post your articles before submitting for publication.  There are a number of journals that accept direct transfer from BioRxiv, so that once you have posted your final draft it is automatically ported andyou do not need to enter all the info into another system.

The reasons I am considering it:

1) The main reason is that it will get our work out quickly, avoiding the delay of peer review etc.  Historically I have not cared too much about this, but I have moved into an area with a lot of interest and would like to establish our lab as a centre for this work.  I am concerned that if I do not get some data/results out soon, we may get skooped.  In essence, this would be calling dibs...   I think this will also be useful to gain further industry support and grant funding.

2) The premise is that you can receive feedback from people before publication, resulting in better papers.  I think this could be useful

3) I dont have too much respect for the peer review process these days.  If it were up to me, I would be tempted to just post my work in places like this for free and let the reader be the judge rather than paying journals to publish my work and be at the whim of peer reviewers who often do very little to improve the work (likely because of the entire publishing system, but that is a whole different rant).  However, given that peer reviewed publications are the metric that my students and I are measure, I must continue on that path but feel this is a good compromise to get things out faster.

Anyway, just thought I would see what people think?



   

[Hibush]

I don't particularly look at BioRxiv because I have excellent access to journals, and have trouble keeping up with just the literature in the good journals. I don't have time to wade through a bunch of unreviewed dross. However, articles in my fields are being read on BioRxiv, so someone is looking.  Publishing in BioRxiv may get you an additional audience.

I have not seen useful comments on articles, ones that can be used for improving the manuscript for publication. It is supposed to be one of the benefits, but seems not to be happening. I'm not surprised since it is so challenging for editors to find reviewers. I have a network of colleagues who will offer valuable pre-submission review and a small number of journals from which I get solid reviewer comments and sound direction from the editor. Those two factors make me more willing to stick with "the system".

Like most active researchers, I try some cool ideas, but the preliminary results indicate it is not a productive research direction. The experiments are well done, and the conclusions are clear. It would be great to be able to publish that work so others can figure out not to repeat exactly that until they have overcome the limitations we ran into. BioRxiv isn't yet an outlet for that sort of thing.

They do offer "Contradictory data". That seems valuable, especially for mouse researchers. There are cases where something gets published in e.g. Science, and then nobody can repeat it. The researchers who assume the result and plow ahead may never know that their results are artifacts. But the ones who go to the trouble of trying to replicate should tell us when it doesn't work. I imagine you don't get too many brownie points for those publications.

BioRxiv does not seem to be part of the citation ecosystem (Google Scholar, ResarchGate, Biosis, Web of Science). That is a big downside now that these systems bring attention to relevant research.

I suspect you might get some of objective 1, will be disappointed on objective 2 unless it is the culture in your subfield, and that this is not the best solution to the issues in objective 3.

[Liquidambar]

I'm in a field where it's common to post preprints on arXiv (https://arxiv.org/), a preprint server originally for physics but now for other fields like math and computer science.  I usually post there at the same time that I submit a paper to a journal.

Pros:
1) The preprint can be linked for people who might want to see the work now instead of waiting for the publication.  (E.g., people listening to my conference talks, potential employers of my students, funding agencies that I hope will appreciate my productivity.)

2) It's a tiny bit of extra publicity.  Anything helps.

I thought I had another pro, but now I've forgotten it.

Cons:
1) If your paper has a major flaw that the reviewers identify, it's embarrassing that the preprint has already been out for months for all the world to see, though of course you can upload a new version once you fix the problems.  I've only had one preprint (of 20+) that I wish we'd waited to post.  It wasn't wrong, but the paper could have been a lot better, and one of our observations had a simple explanation that we failed to notice.

In my field, preprints get cited when the published paper isn't available yet to cite.  I read a decent number of preprints because there's interesting work that isn't published yet.  Usually it's because I heard a conference talk about it, or because some software (Mendeley or Researchgate) decided I might be interested in it, or occasionally because it came up in a web search.  I hope biorXiv will eventually be used in this way.  Actually, I read a biorXiv paper just today because I'm collaborating with the author and extending some of his earlier work that hasn't been published yet.  Rather than e-mailing me a preprint, he just told me to get it from biorXiv.

1) The main reason is that it will get our work out quickly, avoiding the delay of peer review etc.  Historically I have not cared too much about this, but I have moved into an area with a lot of interest and would like to establish our lab as a centre for this work.  I am concerned that if I do not get some data/results out soon, we may get skooped.  In essence, this would be calling dibs...   I think this will also be useful to gain further industry support and grant funding.

2) The premise is that you can receive feedback from people before publication, resulting in better papers.  I think this could be useful

3) I dont have too much respect for the peer review process these days.  If it were up to me, I would be tempted to just post my work in places like this for free and let the reader be the judge rather than paying journals to publish my work and be at the whim of peer reviewers who often do very little to improve the work (likely because of the entire publishing system, but that is a whole different rant).  However, given that peer reviewed publications are the metric that my students and I are measure, I must continue on that path but feel this is a good compromise to get things out faster.

I think (1) is a good reason.  I know an emeritus professor who planned to post all his work on arXiv instead of publishing (as in your (3)), but he keeps collaborating with younger people who need publications, so he keeps submitting to journals anyway.

BioRxiv does not seem to be part of the citation ecosystem (Google Scholar, ResarchGate, Biosis, Web of Science). That is a big downside now that these systems bring attention to relevant research.

Google Scholar and ResearchGate seem to know about arXiv, so hopefully they'll eventually index biorXiv as well.  I don't expect Web of Science ever to index such things, since Web of Science is for published work.

[youllneverwalkalone]

I am not in a field that traditionally publishes pre-prints, but I also see it becoming more and more common.

At the moment, I do not see yet see any value in doing that to be honest. Scooping isn't really an issue for me, the early feedback thing seems more theoretical than real, and I am not yet so senior or disgrunted with the system to do without actual publications.

The part below is attractive though, so if that becomes a standard I'll reconsider.

There are a number of journals that accept direct transfer from BioRxiv, so that once you have posted your final draft it is automatically ported and you do not need to enter all the info into another system.

[polly_mer]

I'm in a field where it's common to post preprints on arXiv (https://arxiv.org/), a preprint server originally for physics but now for other fields like math and computer science.  I usually post there at the same time that I submit a paper to a journal.

arXiv was a lot more useful to me when the publication timeline was several months from submission to acceptance and then a couple months from acceptance to publication.  For many of the topic areas that I read, the time from submission to available as a preprint on the journal host is now under 8 weeks.  What ends up languishing on arXiv are the ones where reviewers are having substantial discussions and the paper is either undergoing significant revision or being shopped around after rejection.  However, a handful of articles from the 1990s in an area where I was publishing a lot 15-20 years ago never were published in another venue and so the arXiv PDF is highly cited as the first paper in the area and continues to be the standard reference.

I've also seen an uptick over the years in some particularly interesting-to-the-public-and-therefore-attractive-to-science-fans-who-don't-have-enough-science-background areas on arXiv where what's posted is just plain wrong and will never be published in a peer-reviewed outlet.  Years ago on the CHE fora we had someone who spent months trying to convince us that the science publication method of 2-3 peer-reviewers for each article to be collected in journals was wrong and the immediately peer-reviewed articles on a server somewhere was the wave of the future.  I'm still not seeing that future where I am, although sharing the data and analysis code on a publicly accessible server after the paper is published is having an uptick.  Consolidating papers through ResearchGate, Google Scholar, or one of a handful of other similar outlets is more common than it was 15 years ago.

[Kron3007]

Yeah, I know the feedback aspect often doesn't happen, but I just read a fairly controversial paper on there that is related to my work that has some somewhat heated debate.  As a result, the author ended up adding more info that they missed, so it does happen.  Perhaps reviewers would have mentioned it too, but now they won't have to.

Anyway, I think I will give it a try and see what we think.  My initial reluctance with these is that I saw a CV from a postdoc listing several nature preceedings (same idea, but nature scrapped it) under publications.  For this approach to work, I think it is important to clearly label them as pre-prints. 

[Liquidambar]

arXiv was a lot more useful to me when the publication timeline was several months from submission to acceptance and then a couple months from acceptance to publication.  For many of the topic areas that I read, the time from submission to available as a preprint on the journal host is now under 8 weeks.

Wow.  My field is still on your old timeline.

Anyway, I think I will give it a try and see what we think.  My initial reluctance with these is that I saw a CV from a postdoc listing several nature preceedings (same idea, but nature scrapped it) under publications.  For this approach to work, I think it is important to clearly label them as pre-prints.

I have a section on my CV for submitted manuscripts, so I just include the arXiv link there when available.  I try to teach my students to do the same, but of course not all of them listen.

[Kron3007]

arXiv was a lot more useful to me when the publication timeline was several months from submission to acceptance and then a couple months from acceptance to publication.  For many of the topic areas that I read, the time from submission to available as a preprint on the journal host is now under 8 weeks.

Wow.  My field is still on your old timeline.

Anyway, I think I will give it a try and see what we think.  My initial reluctance with these is that I saw a CV from a postdoc listing several nature preceedings (same idea, but nature scrapped it) under publications.  For this approach to work, I think it is important to clearly label them as pre-prints.

I have a section on my CV for submitted manuscripts, so I just include the arXiv link there when available.  I try to teach my students to do the same, but of course not all of them listen.

Yeah, my field can still be very slow.  There is a major issue with lack of reviewers, so I have had them come back months after submission only to tell me they have not found reviewers and ask me for more suggestions (this whole practice is crazy in the first place).  Then in cases where it is rejected, which can happen simply based on lack of perceived impact, reviewers' odd preferences, etc, the delay can be long.  This is one of the main reasons I like the idea, at least the information is out there while the process is moving forward (or not).