News

Industry and academic leaders collaborate to launch new publishing platform to improve the reproducibility of preclinical research

F1000Research will launch a new publishing channel that will encourage and facilitate openand transparent publication and discussion of confirmatory and non-confirmatory studies inthe biomedical research sector.

Championed by Amgen’s Senior Vice President for Research, Sasha Kamb, and former Editor-in-Chief of Science and F1000Research International Advisory Board member Bruce Alberts (University of California, San Francisco), the Preclinical Reproducibility and Robustness channel is open for everyone to publish and discuss confirmatory or nonconfirmatory scientific research results.

In an increasingly competitive global environment for funding and employment, research scientists feel pressured to ‘publish or perish’. This heavy burden sees a lack of encouragement to carry out and publish research that tests, confirms or contests preclinical research findings. This means that much follow-on research builds on study data that havenot been confirmed or reproduced, and which inevitably results in significant research wastage.

Rebecca Lawrence, Managing Director, F1000 said: “Until now, there hasn’t been a dedicated place to discuss why some studies can’t be reproduced in a fair and controlled environment. The F1000Research editorial policies and transparent publishing model are well suited to provide such an environment through the provision of all the source data, and open and transparent publication of the whole peer review and post-publication commenting process.”

To address the issue of research reproducibility, in mid-2015 the UK’s Academy of Medical Sciences (AMS), and leading funding bodies for biological and medical research – the Biotechnology and Biological Sciences Research Council (BBSRC), the Medical Research Council (MRC) and the Wellcome Trust – held a symposium to explore the challenges associated with the reliability and reproducibility of biomedical research in the UK, and opportunities for improvements. Dorothy Bishop, Chair of the steering committee for the resulting report, titled Reproducibility and Reliability of Biomedical Research, said: “Tackling the reproducibility issue will need a mixture of ‘top-down’ and ‘bottom-up’ approaches, but a strong community effort is essential to underpin the problems that are felt on the ground by practising scientists.”

Bishop welcomed the new F1000Research platform. “This new channel facilitates many of the strategies that our report concluded as important: open methods, open data, post publication peer review and collaboration.” The Preclinical Reproducibility and Robustness channel will provide an environment for reinforcing the acceptability of being open about the results of researchers who are attempting to assess the robustness of major scientific findings, whether they are confirmatory or not. Amgen will be the first to publish three studies on the Preclinical Reproducibility and Robustness channel, which will encourage global discussion centred on the original research and Amgen’s non-confirmatory studies, and accelerate understanding of the underlying science.

The challenges of reproducibility affect both pharma and academia. Larry Tabak, Principal Deputy Director, National Institutes of Health (NIH), the largest funder of academic biomedical research in the US, said: “Recently the NIH has put considerable effort into improving the reproducibility of the work it funds”. These efforts include developing a training module on enhancing reproducibility of research findings and promoting and ensuring increased transparency for all NIH-funded clinical trials on clinicaltrials.gov. Tabak went on to say “this effort led by Amgen shows there is an appetite from industry to engage in a constructive dialogue about preclinical studies, and this new channel is something we support and encourage our researchers to contribute to.”

Brian Nosek, Executive Director of the Centre for Open Science and one of the lead coordinators of the Reproducibility Project: Cancer Biology, said: “Trying to build on irreproducible results is wasteful and inefficient. This channel offers an opportunity to identify reproducibility challenges early, and dramatically accelerate the accumulation of knowledge and translation of basic science into effective clinical interventions.”

Alberts, a co-founder of the ‘Rescuing Biomedical Research’ project, and formerly president of the National Academy of Sciences, strongly supports this use of F1000Research to enhance the self-correcting nature of science. He said: "Essential for the advance of science is the ability of a scientist to attempt to reproduce the published findings of another scientist and then readily publish his or her results. In recent years, it has become painfully obvious that the biomedical sciences urgently need an improved mechanism for openly and rapidly publishing non-confirming results, in a way that allows contentious issues to be efficiently exposed and discussed by scientists without editorial bias. To me, the F1000Research publishing platform is ideally suited to do this, and I very much hope that the new channel will be widely used for this purpose".

Kamb said: “In discovery research, we dedicate a huge amount of resources to reproduce, and hopefully extend, published studies, and in many instances we can’t. The goal of Amgen and our industry peers is to ensure potential therapies with strong clinical rationales move more quickly through the discovery pipeline. We, like the entire scientific community, depend on fundamental discoveries made around the world and communicated via publications, and we can restore balance to the publication process by removing some of the bias against non-confirming dataset publication. This will improve efficiency and speed in translating basic discoveries into new medicines by decreasing the waste from redundant, non-publicized attempts to build on non-robust findings, and allowing us to focus more on the real breakthroughs.”

Share this article

More services

 

This article is featured in:
Target Identification/ Validation

 

Comment on this article

You must be registered and logged in to leave a comment about this article.