blog

In response to rising concerns about irreproducible science and the lack of somewhere to openly discuss these issues, we recently launched the Preclinical Reproducibility and Robustness Channel. Our aim is to provide a place to publish confirming and non-confirming studies with the full methodologies and underlying data being made available. We want the channel to become the breeding ground for post publication discussion, transparent peer review and open commenting for the scientific community to self-regulate their research interests. Our hope is that this channel could help erode the notion that once an article is published, that work is essentially completed, and start to change the culture of how experimental data is recorded.

This has happened to some extent on the three articles that we launched with, published by the pharmaceutical company Amgen. These articles were all attempts at replicating studies from the literature, and have been openly peer reviewed by experts in the field. The reviewers have given their thoughts on the replications and there has also been interaction from the original authors, as well as comments from other scientists.

Since then, we have also published two very interesting articles that are (at the time of writing) awaiting peer review:

1. A replication study where four individual pharmaceutical labs tried to reproduce the findings of a paper which showed potential for a biological mechanism of ketamine’s anti-depressant action. The labs used a networking forum to discuss the original research and identified the study as being difficult to reproduce. Together they assembled their findings and published their results, disclosing their full methods and data.

2. A validation study from an academic lab that wanted to confirm a previously published finding which suggested milk intake increases presence of bovine miRNAs in human plasma and peripheral blood mononuclear cells.

 We want you to publish your studies too!

At F1000Research, we pride ourselves on transparency and speed, and we encourage the publication of all sound science. We believe that all valid and well conducted research should be published; positive, negative or null, it should be part of the scientific record.  This is very important for scientific progress – and we know that many of you agree with us.

So we encourage you to join us in helping to improve the scientific record by publishing your confirming and non-confirming and submit them to F1000Research. To submit an article to the channel please use the channel submission form.

The majority of the articles we have published in the channel have been categorised as a “Research Note”.  Research Notes are papers that don’t necessarily need a huge introduction or discussion section, they mainly include the relevant context of the work with detailed methods and protocols, and a clear description of the results supported by the source data. These papers therefore shouldn’t take as long to write up and can be published very quickly after acceptance.

By publishing these studies two important things will be achieved:

1. It enables authors to get credit for experimental work that would normally fall by the wayside. Lots of time and effort go into trying to reproduce a study and so to be able publish it provides some tangible recognition.

2. It allows authors to be more certain that they are building upon a more reliable evidence base.

We are currently building up the advisory board for the channel which will include leading researchers in academia and industry and will span a range of disciplines to help guide us as we look to develop and expand this channel, and so if you are reading this and want to be involved please do get in touch.

Finally, below is a comment from Professor Jasper Rine, UC Berkeley on the importance of the PRR channel and the impact that it could have:

“In my experience, studies that provide non-confirmatory results run the gamut from demonstrations that a published result is fundamentally wrong, to demonstrations that a published result is valid in only certain strains or under limited conditions. In either case it is extremely difficult to make people aware of the error, in some cases, or limited robustness in others. Journals are reluctant to publish such studies, and it is often harder to refute a published result that it was for that result to be published. If we all use this new mechanism responsibly, I see the potential for great benefit. I hope we all embrace the “preclinical” part of the channel’s title to include even the most basic aspects of research”