Retraction Watch

Tracking retractions as a window into the scientific process

How does Jeremy Berg plan to address reproducibility in Science?

with 10 comments

Jeremy Berg

Jeremy Berg, via AAAS

The former director of the National Institute of General Medical Sciences at the U.S. National Institutes of Health has a new job. On July 1st, biochemist Jeremy Berg will take the helm as the editor-in-chief of Science. He’s currently the associate senior vice chancellor for science strategy and planning in the health sciences at the University of Pittsburgh. We spoke to him about challenges he’ll be facing in his new role: treating science’s replication problem, boosting transparency, and making papers as widely available as possible.  

You told us in an earlier conversation that diagnosing and treating science’s replication problem is major issue in publishing. Can you give us some specifics about how you plan to address it at Science?

There is considerable activity already underway at Science as Marcia McNutt has been one of the leaders in this arena. See, for example, “Promoting an Open Research Culture”, Science, 348: 1422-1425 (2015). I need to get up to speed regarding what is already underway. The point of my comment is that difficulty in replicating a scientific study can occur for a variety of reasons and the most effective approaches likely depend on the cause.

What are some of the other major issues facing the industry, and how do you plan to address them at Science?

A major issue revolves around finding approaches that make scientific articles and information as widely available as possible while providing business models that can cover the costs of peer review, editing, distribution, and so on. I believe in these principles (wide accessibility and sustainability) and am agnostic about how they are achieved. I look forward to working with others at AAAS who have been worrying about these issues for a long time.

You note in your ASBMB President’s Message that you are pro-transparency, but that “Certainly, some information is sensitive and cannot be shared widely without causing difficulties.” What are some examples of when transparency is not appropriate?

I am certainly very much in favor of transparency. I was referring to information such as personal health information or individual salary information that, in most cases, is legally or ethically protected and sharing would not be appropriate. The same constraint applies to the identity of reviewers if they have agreed to serve under conditions of anonymity.

Do you have any specific ideas for how to boost transparency at Science? 

There are a variety of ideas currently under discussion. Stay tuned.

Last year, Science received criticism for promoting damaging stereotypes of women and trans people; we reported on a letter signed by hundreds of academics. As a white male, you haven’t personally experienced issues that face many minorities — how do you plan to make sure the journal and publisher (AAAS) are inclusive?’

It is certainly true that, as a white male, I have not been subjected to many of the issues facing minorities and women in science. However, I have been involved in diversity issues since before my time at NIGMS and was quite involved with diversity program development and evaluation there. I also served until recently on the National Academy of Sciences Committee on Women in Science, Engineering, and Medicine (http://sites.nationalacademies.org/pga/cwsem/ ).

Since I have not yet started at AAAS, I have not yet had direct experience with the organization from the inside. However, while I was at NIGMS, I worked with AAAS staff in their Education and Human Resources program (http://www.aaas.org/program/education-and-human-resources ). I am optimistic that AAAS is paying attention to inclusiveness issues, but will certainly monitor them myself and tackle any issues that emerge.

Since you have a background in the life sciences, how do you plan to handle issues that arise in the physical sciences?

This is an aspect of my background that is widely misunderstood. My training is largely in inorganic and physical chemistry. My research through my PhD was almost entirely in chemistry. Some of it was biologically motivated but the concepts and techniques were almost entirely from chemistry and physics. My last formal training in biology was my AP biology class in 11th grade and I took a one-semester biochemistry class in graduate school. I am comfortable that I will be able to handle most issues across the spectrum of scientific disciplines covered by Science.

Science uses single-blind peer review, in which the identity of reviewers is kept from the authors. Would you consider going double-blind — where the reviewers don’t know the author’s identities — or even totally blind, where even the editor is in the dark on identities?

I am well aware of much of the science behind unconscious basis and it potential impact. Double-blind review is used widely in my wife’s field [radiology] (although the authors’ identities are often easily guessed). I am certainly open to considering experiments in various forms of blinded (or fully open) peer review.

Did you have any hesitations in accepting the position? If so, what were they?

My major hesitations are logistical. I moved from a wonderful position as Director of the National Institute of General Medical Sciences (NIGMS) at NIH to Pittsburgh five years ago when my wife, Wendie Berg, MD, PhD, had a great opportunity in Pittsburgh. We are well settled in here and she is in the early stages of leading a major clinical trial. I am planning to spend as much time in Pittsburgh as possible while handling my new responsibilities. These geographic issues will present some challenges for me, Wendie, and the AAAS staff.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

Written by Shannon Palus

June 14th, 2016 at 11:30 am

Comments
  • Anonymous June 14, 2016 at 1:06 pm

    “providing business models that can cover the costs of peer review”
    Dr. Berg, do peers not review voluntarily for Science? If yes, then what are these costs you are referrng to?

    • Anonymous2 June 15, 2016 at 8:56 am

      Someone has to coordinate peer reviews and reviewers—I’d imagine he means compensating those folks, paying for any systems that the queue of papers might be stored in, and various everyday logistics of the review system.

      • Jeremy Berg June 15, 2016 at 10:23 am

        Yes, those are the costs to which I was referring.

    • Anonymous June 15, 2016 at 12:26 pm

      So, in other words, those who most guarantee the quality of articles published in Science (i.e., the peer reviewers) are not paid? Therein lies your first major chink in reproducibility (i.e., financially unappreciated professionals). An uninspired peer is less likely to do a thorough revision of the work than a properly compensated one.

  • SB June 15, 2016 at 3:52 am

    I find his comments on transparency rather naive and simplistic to the point of total misunderstanding. Transparency isn’t about releasing everything to everyone – transparency is about being clear and consistent in your processes. It is about ensuring everybody knows why decisions were made, under what process and who made the decision, plus declaring conflicts, and also about ensuring accessibility. Releasing everything to everyone is in fact a great way of not being transparent because it’s actually very easy to hide bad data or damaging information in swathes of other information thus making it very hard for an individual to find what they need, whilst also allowing the organisation to abdicate any responsibility.

    • Jeremy Berg June 15, 2016 at 10:22 am

      Thanks for your comment. I agree that being open and clear about process is important. I was responding to a question about what information it might not be helpful to release.

      With regard to process, I described the NIGMS funding decision process as clearly as I could on the NIGMS Feedback Loop blog which I started. See https://loop.nigms.nih.gov/2011/01/the-funding-decision-process/ and other posts. I plan to continue such openness about process and policy in my new position.

    • Bobo June 15, 2016 at 10:30 am

      No, full data release (including the code used for analysis) is a critical part of transparency.

      It’s not an “either-or” thing when it comes to “either you have to describe and justify your analyses clearly or you have to release your data/code”. You have to do both.

  • PDavies June 15, 2016 at 5:10 pm

    The Journal of Biological Chemistry (an ASBMB publication) is terrible about transparency. It does not retract articles with image manipulation and plagiarism and when there is a correction or retraction there is very little information on the rationale. JB should have cleaned up JBC before becoming the editor of Science. And will he finally have the Science arsenic bacteria paper retracted?

    • Jeremy Berg June 15, 2016 at 7:27 pm

      If you have specific concerns, contact Lila Gierasch who became Editor-in-Chief at JBC relatively recently.

    • Anonymous June 16, 2016 at 12:47 am

      PDavies, this is a very pertinent comment. And it raises the question: should former EICs or editors be held responsible for the literature they oversaw or approved before moving on to a new journal or publisher, or after retiring?

  • Post a comment

    Threaded commenting powered by interconnect/it code.