Discovered oxygen-sensing HIF-PH enzymes, essential for the 2019 Nobel Prize. Focus on sound science practices and combatting misconduct. Tweets are my own.
@gaetormom
I grew up in the grey communist Romania, the Iron curtain fell when I was 23. The 18 yo me would have dismissed any description of my life as a 50 year old US scientist as a hallucination. The message would convey to myself? “Enjoy an Amazing Ride” ( ok, 4 words technically)
Dear all, as promised I am attaching a very, very rough “manifesto” draft, with the goal to initiate a real movement against science misconduct on multiple fronts. I apologize in advance about the format ( for some reason I could not generate a thread). Those among you who need
2006 paper on cause of Alzheimer’s is the field's 4th most cited paper ever. But the author faked the data and it was later retracted. This wasted many years & $287+ million in funding as the field chased a dead end. Scientific fraud must be criminalized.
I said it numerous times and I repeat it. The deluge of manuscripts submitted to journals indexed by PubMed has long ago surpassed the review capacity of our scientific community. Not enough qualified people and no incentives to spend time that you don’t have.
A few necessary
It’s outrageous that people who have a long history of fraudulent papers have not suffered ANY real consequences. Worse, they keep having open doors to publish in “big”journals, receive enormous salaries invited to give keynote talks, can still get public funding. THIS MUST STOP!
@slavovLab
Nikolai, this is a potentially major finding. With the obvious caveat, and I am sure you’ll be the first to agree, that it needs to be independently replicated. Couple of quick questions: 1) For a given protein, out of 1000 molecules, how many would depart from the canonical
The pressure for productivity is killing reliable research. We have the same brains as in 1970, but among us are people who run immense labs, publish 10 times more, each paper is far more complex. Such labs are disproportionately represented in “high profile” journals and as such
@TheRealBuzz
People who remember July 1969 should feel privileged to have witnessed arguably the most significant moment in human history. For as long as our civilization endures, Moon Landing will be a historic watershed, as other events in our history books gradually fade into oblivion.
Musings on the recent high profile science misconduct cases from Harvard and elsewhere. Researcher X publishes several “high impact papers” and as a result secures big grants then a coveted position at a top institution. Years later, it turns out that several (if not most) of
AI-generated diagram, errors lead to retraction – RW. One of the bazillion examples of just how “porous” scientific literature has become. There is hardly any filter protecting our scientific record from blatant BS, self-serving exaggerations and such.
@MicrobiomDigest
@PubPeer
Journals should stop accepting so easily “replacement” figures and issue an erratum or corrigendum. These are likely to be as fraudulent as the previous ones, it’s just that they do not contain overlaps.
these papers are based on manipulated and/or faulty data.
Papers are retracted, but the career goes on, give or take a polite slat on the wrist. What message does this send to young researchers? How solid are the foundations of science these days?
I keep remembering the words of Carl Sagan about science being our candle in the dark. In absence of its light we are headed for the abyss. Worse still, there are many famous academics who are holding their own “light” of pseudoscience for the unsuspecting public. People with
Dear likeminded friends, as I was alluding recently, I am planning on starting a society/foundation with the preliminary name Agora, dedicated to fighting pseudoscience. At this point, everything is open for discussions. Will return shortly with objectives, thoughts about funding
@DavidSteensma
One of the many pieces of evidence that putting the equal sign between the IF of journal and the value of a paper is a no no . Father time lifts a good result from the noise and hype.
Donors have poured billions into research. They should consider supporting the cause of science cleanup by funding sleuths. These people work on tiny Patreon donations and contribute more to science than most famous institutions. How can we reach out to philanthropists and
The (very strong) case for criminalizing scientific misconduct · I might add corruption of students and defective mentorship, which is particularly bad for teaching institutions of these stature.
@MicrobiomDigest
What do Nature’s editors argue is a reasonable upper limit of “reviewability”? 60 supplements ? Or maybe 100? We should come to our senses and accept the obvious: the race for Nature papers has gone beyond ridiculous. A paper in 1950, 1970, heck even 2000, could be called that, a
@rlmcelreath
@GustavNilsonne
As I argued before, repetitive impactful science fraud should be be treated as felony. The taxpayers cannot be robbed indefinitely under the tutelage of universities. There must be accountability, otherwise our children will inherit noise rather than knowledge.
Nature/Science papers these days: you convince 3-4 people who have limited access to your original data, even less bandwidth and broad expertise. Then that’s it, it’s a “fact” , you get grants and you promote. Ten years later it’s more likely your stuff has not been reproduced,
Many years ago Upton Sinclair anticipated the current crisis in academic life science: “It is difficult to get a man to understand something, when his salary depends on his not understanding it.” This crisis is worsened by the absence of specific laws and institutions that can
Scientific Fraud Is Slippery to Catch—but Easier to Combat | WIRED. Well, I agree with a lot of the content but the title conveys the wrong message. Science fraud is very hard to combat. The academic & publishing system makes it virtually impossible.
“Unfortunately, when we contact authors about concerns that have been raised about their Science papers, we are often met with defensiveness and denial. That needs to change”
@MicrobiomDigest
@Thatsregrettab1
@DrNancyOlivieri
The quality of science is eroded with every grant given to dysfunctional or fraudulent teams, with every paper accepted by big journals from such teams despite mountains of PubPeer evidence. With every pompous press release by their home institution, with every platform given to
" … the public’s trust in science has been undermined when scientists with large public platforms have failed to state strongly enough that their pronouncements are based on science that remains in flux," writes H. Holden Thorp in a new
#ScienceEditorial
.
Shouldn’t the
@NIH
have a “reproducibility office”? Essentially a lab that repeats key experiments in “just published” high profile new papers? Prioritize “hyper productive” groups, sensational claims, labs with a “history”. Even if cost> $100m, now we waste bn/yr in bad data.
The magnitude of scientific fraud is orders of magnitude higher than currently presented. If we had 1000 x
@MicrobiomDigest
the number of problematic papers would be commensurately higher.
What
@MicrobiomDigest
tweets regularly scares me. There is no way I could ever even vaguely guess any of these things as a reviewer/reader. If image fraud is so permeating, numbers (which seem so much easier to doctor) should be tempered with on scales I don't want to think about
Dear all, here is the link to the 2nd draft of "AGORA manifesto". Many, many thanks to dozens of people who provided valuable input (I tried to incorporate as much as I could) based on valuable public input. Below is the link to the Google doc, please RT as much as you can. Based
@ArchitGhosh1
@OdedRechavi
Agree. RT-PCR works well for RNAseq validation when fold changes are dramatic and the control gene is stable in that context. However, many biological responses involve subtle (but coordinated) expression changes and various stresses affect control genes (e. g. hypoxia - GAPDH).
The 7 biggest problems facing science, according to 270 scientists. This was written in 2016. The problem has gotten much worse since then and nothing has been done to prevent it.
This one is the most outrageous of all, all manners of faked figures in most of his papers , going back to the days when he was doing the experiments
Hats off,
@MicrobiomDigest
@peterrhague
The psychological consequences of long term separation from Earth, friends, familiar environment. I am more optimistic about solving the engineering obstacles than I am about addressing the challenges of human mind severed from its home.
This is an essential part of the solution: people should be afraid to commit fraud. The absence of real consequences is tantamount to incentivizing fraud. Carrot without stick.
The word format, please DM me. Please also let me know what might be the best “place” for us to have a detailed discussion. The only “rule” at this point is polite, constructive conversation. Please RT and diseminate, this is the type of discussion that requires open minds from
@talalrahwan
@Nature
I remember the times (around 2000) when I was a believer in the relevance of metrics. Given enough time and enough wrong incentives, humans find ways to pervert any concept, no matter how sound.
In 2015 Newcastle scientist Professor Paul Dent made cancer and ebola breakthroughs
Unfortunately,
@MicrobiomDigest
@Thatsregrettab1
and others don’t want this to happen. They keep making a big deal out of a few manipulated figures (a few dozens or
@MicrobiomDigest
@BostonGlobe
Lost in the article are the countless cases of manipulated figures collected on PubPeer. The number of figures that were “accelerated” to satisfy / impress reviewers is shocking. And it may be only the tip of the iceberg (correct, Dr. Bik?).
@emilyakopp
An agency with a $50bn/yr budget that does not have a investigative branch?…Perhaps this is why it keeps funding folks with a lot of fraudulent papers and never holds them accountable
dear all, as many of you suggested, I have created a google doc for public comments.
Please circulate and let's take it to the next level.
Much appreciated.
@kareem_carr
It took a pandemic to bring Dr. Kariko’s discovery to light. Imagine for a second that Covid19 never happened: mRNA vaccines would still be under development in a biotech company. Dr. Kariko would still be unknown to the public, UPenn would not be bragging about the 2023 Nobel.
When science is done correctly:
1. Most results should be negative.
2. Most efforts should be spent to falsify promising data (i.e. prove them wrong).
Alas, the system is geared towards showing that everything works, The result? Noise rather than reproducible usable data.
“I do not believe that the institution has addressed this appropriately,”says DavidSanders,a molecular virologist
@LifeAtPurdue
& 1 of the signers of the letter.“It’s unbelievable that this is not grounds for more severe consequences.”
#research
#integrity
@cshperspectives
@TepavcevicVanja
The good scenario: Other journals follow suit. IF does not count anymore. One can subsequently add data to the study until it becomes a solid body of work, with community input.
@eLife
@NIH
(The bad scenario: eLife IF “collapses”, papers don’t count much for grants…Hope not)
Reposting this, with an intent to reach as many interested people as possible. During the past decade I’ve spent close to 10k hours learning how to integrate genomic datasets from dozens of public databases. At this point I can generate “literature-agnostic “portraits” of
Why are journals reluctant to retract? EICs know the magnitude of the problem (otherwise they would be plainly stupid). They are aware that without fakes a good chunk of their customers/authors vanish. EICs are under pressure to grow the business. R/T if you agree or comment.
Few random thoughts about making papers reliable. 1) Huge labs and high productivity should be prevented, not rewarded. 2. The standard for indexing journals should go up, way up. At least half of the journals should be delisted. There is simply not enough credible expertise in
A reason why this ocean of junk gets published? Editors in chief greedy to keep their paid position accept anything that superficially looks like a paper. The pace at which legit science can realistically be generated is too slow for the ocean of journals.
Academic publisher Elsevier's profit margin compared to Apple, Google, and Microsoft
Apple: 28%
Google: 25%
Microsoft: 34%
Elsevier: 37% with a revenue of $3.9 billion.
Elsevier's payment to academic authors and reviewers: $0
re: Scopus-indexed Journals “Every year, Scopus receives more than 3,500 NEW journal titles proposed for inclusion in its database”. I am at a loss for words. Who reviews these “journals”? How much trustworthy information is generated every year?
“More is not better: the developing crisis of scientific publishing”. In fact, we can safely say “More is worse!” Too many journals, too many papers, too few reviewers with adequate competence, time and impartiality.
“THE FOUR FORMER lab members who allege Zlokovic pushed them and others to manipulate data paint a picture of a pressure cooker environment in which their boss expected new data almost every week, always in line with his hypotheses”.
Bullying to generate “science”.
If people brag about H index, how about an R-index, or PubPeer index? Why only the “good”? Mind you, what’s on PubPeer is most likely a fraction of the dark side. Also, at what point will “hyper productivity” become counted as a negative?
Show of hands: how many of you have heard about a dean/department chair mention PubPeer and prioritize science integrity? All care about major grants coming in and papers in high impact journals. When senior faculty are caught with many manipulated figures, just crickets.
@aemonten
The “mechanistic” part of a paper is often a desperate response to reviewers. The higher the journal, the higher the pressure for details. The result ? Forced connections that are rarely confirmed or pursued even by the lab of origin. Better no mechanism than a bogus mechanism.
@Thatsregrettab1
@SpringerNature
Scenario 1: Joe Sucker submits a paper, gets the reviews and it’s clear that without experiment X asked by Reviewer 2 the paper is never going to get in. Tries a few times, but nature does not “cooperate”, X does not fit the story. He gives up disappointed. Same scenario, but Joe
@catecadell
The protesters, many completely exposed to surveillance cameras, are facing an enormous risk. Can’t say enough about their courage, they represent China’s true hope for a better future.
@CharlesMBrenner
Hype at the highest level spread to the general public by media outlets. I wonder if he believes any of it himself? Has anyone been able to replicate his claims ?
“ every time we pay a $3000 article processing charge, only $1800 supports the publishing process, while the remaining $1200 goes directly to Elsevier shareholders”. I guess not much left for detecting fraudulent figures, which should be their top priority
During the past decade I’ve spent close to 10k hours learning how to integrate genomic datasets from dozens of sources to generate “literature-agnostic “portraits” of virtually any coding gene, noncoding RNA or regulatory element. In a nutshell, I generate figures that predict
Absolutely. Especially when there was a clear evidence of image manipulation. Why would the authors resort to such thing is a “correct” figure was available in the first place?
I can’t imagine any reasonable excuse from the authors.
One assumption is that regular taxpayers cannot grasp the content of scientific papers and for most part this is true. HOWEVER: an average individual should be able to understand figure duplication. Nobel laureates publishing papers with duplicated figures means we are in deep …
One of the deepest questions that should be part of any discussion with someone harboring an opposing view: “what would it take to change your mind?” We should also practice asking such question to ourselves. Being open to change opinions, no matter how deeply rooted, based on
Planck's principle: “scientific change does not occur because individual scientists change their mind, but rather that successive generations of scientists have different views”. In other words, long term progress would be impossible if the famous and powerful lived forever.
One of the questions asked in this article: “The question of whether research fraud should be a crime”. There should be meaningful consequences for those perpetrating fraudulent science and in some cases yes, such acts should be prosecuted as crimes.
Join us for a timely and hopefully fascinating discussion. Without a doubt one of the most important, rapidly growing yet largely ignored problem of our times. Questionable science practices, pseudoscience, fraudulent & erroneous science, whatever you want to call it. It’s all
How the Karolinska protected Paolo Macchiarini — and whistleblowers paid the price – “Accompanied by members of the Nobel Committee, Hamsten disavowed Gerdin’s external inquiry and proclaimed his confidence in Macchiarini”….
23 years passed between my discovery of enzymatic HIF prolyl hydroxylation (2001) to the first FDA approval of a drug that exploits this mechanism (yesterday). Time flies…
100 Greatest Mathematicians of all Time! ✍️
1. Isaac Newton
2. Archimedes
3. Carl F. Gauss
4. Leonhard Euler
5. Bernhard Riemann
6. David Hilbert
7. Joseph-Louis Lagrange
8. Euclid of Alexandria
9. Alexandre Grothendieck
10. Gottfried W. Leibniz
11. John von Neumann
12. Henri