Open science and transparency serve as antidotes to the epidemic of flawed and fraudulent research. And the journey toward open science begins with preprinting.
This post is a celebration of bioRxiv’s 10th anniversary.
Science is awesome
Science predicts outcomes. It’s that simple.
Consider molecular biology: you mix a restriction enzyme with some DNA, wait a few minutes, and bang, the enzyme cuts precisely where it should. You design a small guide RNA to match a specific region of a genome, transform it into the cell with the Cas9 enzyme, and voilà, the CRISPR/Cas9 molecular scissor cuts exactly where intended and you end up with genome edits at that precise location.
Examples like these abound. Antibiotics annihilate bacteria, thwarting infections. Vaccines reduce the risk of viral infections. Plant disease resistance genes confer immunity to pathogens and pests.
It’s reliable, reproducible — anything but magic, though it often feels that way. That’s why we’re drawn to science. It’s why many of us are utterly hooked and captivated by science. It’s simply awesome.
Cheats and fraudsters
The very essence of science lies in its ability to deliver reproducible and predictable outcomes. That’s why it’s deeply disturbing to regularly encounter tales of cheats and fraudsters within our scientific community. The catalyst for revisiting this topic was Chris Said’s X/Twitter thread, detailing how several prominent Alzheimer’s researchers are facing credible accusations of fraud. One instance even led to drug trials involving a thousand volunteers — a shocking revelation, to say the least.
In the words of Aaron Charlton, “It’s interesting to see how all the fraudsters converge on a disease with no cure and no hope for effective treatment on the horizon.”
Aaron hints at the possibility that some of these cases may involve calculated actions. Does it really matter? It’s unsettling to recognize that instances of flawed or, worse, fraudulent science are unfortunately not isolated events but rather a daily occurrence.
Regrettably, addressing these issues is easier said than done. Often, the questionable work has already found its way into seemingly credible scientific journals. Should we halt our daily pursuits and research investigations to invest hours in reporting bad and fake science on platforms like PubPeer? The sheer volume of such cases suggests it would necessitate a full-time commitment.
Tobias Straub rightly points out the deafening silence that often shrouds these matters. The intricate web of incentives within academia, coupled with the inherently conflicted peer-review system, leaves few rewards for those who dare to expose bad or fraudulent science. This vacuum has been filled, to some extent, by the likes of Elizabeth Bik and Leonid Schneider — dedicated sleuths in the realm of scientific integrity. Their blogs Science Integrity Digest and For Better Science make for incredibly depressing readings. And that’s most likely only the tip of the iceberg. Given this mess, shouldn’t funders and academics do more to deal with this science crisis?
Have we lost the plot?
Let me emphasize it once more: good science is nothing short of magnificent. It’s awe-inspiring, reproducible, and yields highly predictable outcomes. Just the thought of the profound impact that robust, solid knowledge can have gives me goose bumps.
My dream, and indeed the dream of every scientist, should be to unearth knowledge of this caliber — knowledge that empowers us to foresee future events, knowledge that stands the test of time and bestows upon us the ability to make dependable predictions. Only then can we retire with the satisfaction of having accomplished our mission.
Yet, what I often witness is a relentless pursuit of journal publications, the propagation of wishy-washy science, and even a questioning of the very essence of science itself. It’s as if some have lost sight of why we became scientists in the first place.
A colleague once claimed that scientific certainties would always be overturned by new discoveries. ALWAYS overturned? That’s nonsense. The restriction enzyme EcoRI will always cut G/AATTC, irrespective of any new scientific revelations. I don’t expect Kepler’s laws of planetary motion to be overturned anytime soon — they’ve been with us since the early 17th century. Our understanding may evolve and expand, but outright overturn? Sometimes, yes, but not ALWAYS.
If you embrace this view of the world, then perhaps you shouldn’t fly in airplanes, just in case the science of aerodynamics gets overturned while you’re airborne.
We are scientists, not nihilists. We firmly believe that an objective reality exists, waiting to be unraveled. Our duty is to unveil that reality and contribute to the ever-growing tapestry of human knowledge in order to make robust predictions.
Open science to the rescue
Open science and transparency stand as antidotes to the growing epidemic of flawed research. As we celebrate the 10th anniversary of the preprint server bioRxiv this week, it’s crucial to remember that a commitment to open science starts with embracing preprints in all their forms, not just full papers but also posters, datasets and methods.
The fact is, those who publish manipulated or subpar science tend to shy away from preprints. Why? Because they are well aware that open science invites the scrutiny of the entire scientific community, exposing the flaws in their work. Their motivation for publishing is driven not by a desire to share knowledge but by the pursuit of career perks and advancements associated with listing journal publications on their CVs. The last thing they want is for their papers to be meticulously examined or for others to attempt to replicate their experiments or build upon their findings. Obtaining functional biomaterial from them is generally an exercise in futility.
They claim that preprints are useless because they aren’t peer-reviewed even though they know very well that anonymous journal peer-review can be gamed and flawed. You believe in peer-review? Then preprint your papers so we can all peer-review them.
Preprint or else
Open science starts with preprinting. I understand this might ruffle some feathers, but if you’re a biologist in 2023, a full decade after the launch of bioRxiv, and you still don’t preprint your work, I’m inclined to question the quality of your science. Preprinting requires courage, and this courage lies at the core of scientific inquiry. You make a discovery, share it with the world, and await the scrutiny of your peers in the anticipation that they will replicate and expand upon your findings. This, indeed, is the ultimate measure of success in science.
Preprints are a crucial step in that direction: can your science be replicated. It takes guts to preprint, but in fact it takes guts to be a scientist, period.
That’s precisely why I’m frustrated with funding agencies for their slow adoption of preprints. They should either directly mandate preprints (Plan U) or encourage them indirectly, as HHMI has done. Sharing new scientific knowledge months, if not years, before it’s published in journals benefits not only society but also fosters the principles of open science. It’s the first stride toward creating a publishing ecosystem that is transparent and open to thorough scrutiny.
If you want your science and your papers to be taken seriously, then preprint. Otherwise, we must assume that you have something to hide and unwilling to have your work withstand peer-review and open scrutiny.
Acknowledgements
I’m grateful to the colleagues whose tweets inspired this article. I dedicate this post to the bioRxiv Team and wish them the best for their 10 year anniversary. This post was written with assistance from ChatGPT.
This article is available on a CC-BY license via Zenodo. Cite as: Kamoun, S. (2023). Scientists, have we lost the plot?. Zenodo. https://doi.org/10.5281/zenodo.10137829
Suggested readings
