This is a story of a very strange and scientifically confusing stealth correction for a 10 year old paper in Science, but let’s first bring some perspective.
Just when 2020 began, the Nobel prize winner and US-American chemistry professor Frances Arnold retracted her Science paper, published in the same year she received her Nobel prize: 2018. She announced it in a tweet:
It seems, the first author Inha Cho neglected to perform the actual experiments, because the retraction notice said:
“After publication of the Report “Site-selective enzymatic C‒H amidation for synthesis of diverse lactams” (1), efforts to reproduce the work showed that the enzymes do not catalyze the reactions with the activities and selectivities claimed. Careful examination of the first author’s lab notebook then revealed missing contemporaneous entries and raw data for key experiments. The authors are therefore retracting the paper.”
Every scientist can become victim of coauthor’s research misconduct, and it takes exceptional decency and research integrity to retract your own paper voluntarily. Arnold should be applauded for that, as should be other scientists who retracted their works without being asked or, as it almost always happens, forced to.
Instead, the media jumped on the story, BBC even addressed the academic as “Ms Arnold” with the undertones of “Woman gets Nobel, fails in Science”, which presumably will be used by certain characters as argument for why women so rarely get Nobel Prize in chemistry, physics or physiology / medicine.
Here a female Nobelist took responsibility for her lab member’s likely research misconduct which she did overlook initially, but eventually uncovered herself, and then acted most exemplary. Yet no media was interested to compare her actions to those of certain other Nobelists.
How about Louis Ignarro, who coauthored many papers with massively fake data, while peddling phony health supplements and magic pomegranate juice? How about Sir Martin Evans, who teamed up with some notorious liar and struck-off dentist to offer miracle cures to heart attack patients using fraudulently-patented cells of very questionable origin, free of any scientific background? Both Ignarro and Evans are Nobel Prize winners in physiology or medicine and highly respected by every science journalist. Which brings us to another case, that of chemistry Nobel prize winner Sir Fraser Stoddart and a paper he and his coauthors recently stealthily corrected in Science. Nobody writes about that either, because these are all successful men of science after all.
This is the Science paper, from the lab of Omar Yaghi in Berkeley, University of California:
Qiaowei Li, Wenyu Zhang , Ognjen S Miljanic, Chi-Hau Sue , Yan-Li Zhao , Lihua Liu , Carolyn B Knobler , J. F. Stoddart, Omar M Yaghi Docking in metal-organic frameworks Science (2009) doi: 10.1126/science.1175441
On 14 March 2019, a comment was posted on PubPeer:
“Figures 4A and D show MOF-1001 before and after docking of PQT2+, a single crystal–to–single crystal transformation.
The process from A to D is described as “MOF-1001 crystals were introduced into a saturated solution of PQT·2PF6 in acetone”. The supplementary video also shows that shaking is involved. The bottom part of Figure 4A and D shows that the crystals have moved. I thus wonder how the crystals can be unaffected by the washing process, as they do not move at all between the two pictures.
Are Figures 4A and D for illustrative purposes? Were crystals fixed onto a substrate for the purpose of those pictures? I welcome clarification on this experimental aspect.“
Indeed, it is very confusing. The article was accompanied with an online video showing how a large volume of paraquat solution was added to a flask with MOF-10001 crystals, which was then shaken. The added fluid at least doubled the total volume (also the photos in Figure 4A and D prove it), which is bound to have caused much turbulence, with a mystery remaining how this would have left all crystals unmoved. And the video even showed the flask being shaken. Maybe the authors made a mistake, and accidentally used a colour-adjusted placeholder picture twice?
On 31 August 2019, an anonymous reader approached the journal Science with their concerns:
“I thus wonder how the crystals in the top of Figures 4A and 4D can be unaffected by the washing process, as they do not move at all between the two pictures. There is, in fact, not a single pixel displacement of any of the crystals, between the uncomplexed and complexed crystals in Figures 4A and 4D. I do not understand how that could be physically possible, and wonder: Are Figures 4A and D for illustrative purposes? Were crystals fixed onto a substrate for the purpose of those pictures?“
Science editor Jake Yeston immediately replied:
“This paper was published too long ago for us to pursue a formal data request. You are free to contact the authors directly.“
The reader forwarded this exchange to me, and I wrote to Yeston, asking why Science deems a 10 year old paper suddenly too old to bother about its data integrity. Yeston replied citing “poor judgment on my side” and explaining:
“After further discussion, we have contacted the authors to request an explanation for the concern expressed in the PubPeer comment, notwithstanding the time elapsed.”
Common sense would let you expect an upcoming corrigendum, where the authors would apologise for an inadvertent mistake, replace the images in Figure 4 and remind everyone that all conclusions remain unaffected. Instead, something much stranger happened.
In late October 2019, Yeston and the then-Editor-in-Chief Jeremy Berg announced to issue a clarifying correction. Because, according to the editors, the two images were actually perfectly correct as the authors explained. Yeston clarified:
“Just to be clear, their claim is that the sample was under the microscope the entire time, the same sample of crystals in a solvent suspension. They introduced the paraquat reactant slowly and carefully so as not to perturb the crystals. That’s what they clarify is being shown in the image—not a dry sample of crystals that they spilled out of a vial twice in order to take two different photographs. I was initially confused too about how the latter interpretation could be physically possible, but I think the clarified experiment does indeed seem feasible. I corresponded and talked with three of the authors extensively, and I believe they made a thorough case for the integrity of the experiment.“
The anonymous reader remained unconvinced:
“To be frank, that you can introduce a reactant in the solution and the image does not move at all, not one of the crystals by a single pixel, is a feat I would not be able to achieve, nor anyone in my group.
I don’t doubt the science reported in the paper, and the authors provide ample evidence and characterization of the compounds obtained. But I believe they cut corners on this particular subfigure. I guess everyone will be able to draw their own conclusion once all this is clarified.“
On 5 November 2019, Science issued a stealth correction. It has no DOI, it is not available as separate article and it can not be googled or accessed freely. There is no mention at the paywalled abstract page that the article was corrected. Not really the way a Corrigendum is supposed to be published, but the journal told me that they see if differently.
To get to the correction, you must log in, either via Science subscription or a free AAAS user account, which lets you read papers older than 1 year. Once you did that, you will probably search for the correction notice in the full text, but there is nothing. Instead, you must scroll to the bottom of the full text page to find a link to Supplemental Online Material. It is a new supplementary file, with some new revelations in it. After following that link, you will find, among other things, this text:
“Correction (5 November 2019): After the publication of this Report, the authors have provided a supplementary addendum to incorporate further technical details on the experiment monitoring the paraquat inclusion process in MOF-1001 without disturbing the crystals in Section S1. Figure S0 and accompanying paragraphs explain the experimental setup.
The original version is accessible here.”
To find the new addendum, you must download the updated supplement file and proceed to page 20. There, you will find a new section, and now we see it was not just a small spot where the crystals did not move, no, it was the entire bottom of the flask. A large field of crystals there remained perfectly immobile after a large volume of fluid was gently added, as if they were all glued. Except that they were not:
“Monitoring paraquat Inclusion without Disturbing Crystals:
To prove that the MOF pseudorotaxanes were obtained in a single-crystal to single-crystal transformation, rather than the dissolution and recrystallization manner, several MOF-1001 single crystals were placed at the bottom of a 20-mLvial in 5 mL acetone solvent, and drops of paraquat solution were added gently using pipet without disturbing the crystals. The whole process was monitored under optical microscope, and snapshots were taken at 0, 1, 2, 25, and 30 mins after adding paraquat. Partial pictures of the crystals at 0 and 30 mins were presented in the upper panels of Figs. 4A and 4D in the main text.”
Now, it may sound like kindergarten-like behaviour, but no, this is elite science on the highest level imaginable, Nobel Prize and all. It is worth noting that nowhere in the main text did the authors discuss the relevance of their experimental setup, neither previously or even now. It most obviously only became somewhat important after the journal reached out to the authors 10 years later regarding the PubPeer comment. The word “microscope” was not mentioned once before, not in the main text nor in the old supplement, which never bothered to explain the impressive paraquat addition technology in this regard. A technology which we now learn manages to keep every single crystal in its place despite these crystals not having been immobilised in any way as the authors admit.
And as for “drops of paraquat solution were added gently using pipet without disturbing the crystals“, how does one do that if the volume of paraquat solution added is (according to all information in the paper) at least twice the volume of the MOF-1001 crystal suspension? In the normal world, adding even one drop would have moved some crystals. But not here.
The only reasonable answer is: magic. Or the Infinite Improbability Drive. This is exactly why the Nobelist Fraser Stoddart will never retract a paper, while Frances Arnold retracted hers.
The new Science Editor-in-Chief Holden Thorp announced maximum transparency in corrections and retractions with his new 2020 editorial which accompanied the Arnold retraction. But for Yaghi and Stoddart, the journal decided against too much openness. I understand it, the correction is too embarrassing and plain weird, best to be hidden where nobody will find it. Reminds me of the Hitchhiker’s Guide to The Galaxy:
“But the plans were on display…”
“On display? I eventually had to go down to the cellar to find them.”
“That’s the display department.”
“With a flashlight.”
“Ah, well, the lights had probably gone.”
“So had the stairs.”
“But look, you found the notice, didn’t you?”
“Yes,” said Arthur, “yes I did. It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying ‘Beware of the Leopard.”
If such magic crystal technology really does exist in this universe, Stoddart deserves another Nobel Prize. To the very least, he and Yaghi should be lauded by some Vogon poetry.
Readers pointed me towards this interesting article in the ACS magazine C&EN from 2017. Right after he won the Nobel Prize, Stoddart, together with a business partner, set up a business called PanaceaNano, offering nanotechnology-based skin ointments: “Noble” line of antiaging cosmetics including a $524 formula described as an “anti-wrinkle repair” night cream”. This is how it works, as advertised by ACS:
“The nano-cubes are made of carbohydrate molecules known as cyclodextrins. The cubes, of various sizes and shapes, release ingredients such as vitamins and peptides onto the skin “at predefined times with molecular precision,” according to the Noble skin care website. […]
Stoddart himself declares, “It is an exciting moment to witness the birth of commercial products that improve the quality of life of people based on renewable, safe, organic, biodegradable functional nanomaterials.”
Anyone still mistrustful towards Sir Fraser’s scientific claims?
If you are interested to support my work, you can leave here a small tip of $5. Or several of small tips, just increase the amount as you like (2x=€10; 5x=€25). Your generous patronage of my journalism, however small it appears to you, will greatly help me with my legal costs.
The whole scientific system is putrid up to the Nobel Prize level since long ago and needs deep reforms since publication transparency including publication of original data, public and named peer-review of articles and grants, the definition of intellectual property and clear and transparent rules to access an academic career.
A lot of problems would be solved if there was not such a big income and job security jump from the people who generate data (grad students, post docs) and the ones that supervise the data generators. Give grad students and post-docs decent income and job security (like a faculty member), there will be less incentive to be like “Mike”, who has totally distorted his ethical sense to see to it that he will make a lot of money as a faculty member, even if it means cheating.
In other words, the system is dead. Some good data will be generated, but only by those who have no ambition and can withstand the opprobrium that come with that. And even these people will be squeezed out by the “productivity” of the cheaters.
As a small note: there are many PIs who are not faculty members, and rely on 3-5 years terms of extramural funding. I think the majority of PIs are actually not appointed as professors in Germany.
Could they provide at least 2 independent images for forensic checking?
When their president no longer has to deal with what’s true, Nobel prize winners don’t feel obligated to do so.
Adding drops to a flask filled with a solution and with crystals lying on the bottom, without moving said crystals, seems feasible to me. That does not negate the criticisms on the intial reaction of Science or how (in)accessible the correction is, but what is reported here is not enough evidence for fraud, just an incomplete description on how the experiment was carried out.
The Baltimore case . a throwback.to decades back
A century and a half ago, Charles Darwin could work out his theory of evolution by natural selection more or less in the privacy of his own country house. ”Big science” has long since ceased to bear any resemblance to a leisurely gentleman’s pursuit, however; it is ferociously competitive, it has an insatiable thirst for public funds, and it is now everyone’s concern. How poorly we know how to handle this Frankenstein’s monster comes across in ”The Baltimore Case,” a splendid study of a major contemporary scientific scandal by the eminent historian Daniel J. Kevles.
Science on trial
Imanishi-Kari case ends, but debate on scientific conduct continues
Kelves admitted early on his in book that he has a bias favoring Baltimore, which is something an “eminent” science historian would not do. How about “biased” science historian, especially if you spend time in the same school as the one you are reporting on? Brown-nosing distorting the truth, so if you want to understand what really happened, you will probably need to read other sources as well. Entertaining book, though, as it paints, in exquisite detail, a understandable portrait of a dysfunctional (Imanshi-Kari) lab.
If the crystals are grown in the vial, then after you wash off the loose ones, the ones that are left will often not move even if you vigorously shake them inside a solution. This is pretty common actually. If the crystals are placed there and they are lose, then even adding solvent very carefully will move some of them and it’s a really bad excuse to use.
That said, like the person who made the initial complaint said, it does not matter for the basic chemistry that much that they manipulated a vanity image. In the current case, the image gives no useful scientific information and you’re depending on the X-ray data.
And another thing, I highly doubt the ‘Nobel winner’ looked at this paper for longer than one hour. This is probably mostly on Omar Yaghi’s lab since these are MOFs and Stoddard’s group focused on different things. But since he probably will win the Nobel anyways sometime in the future (well, I give him 30% odds), then the rhetoric might still count.
Your treatment of F. Arnold is more than kind and I can sense some double standards (compared to Aguzzi, for example). The way the above case stories are presented looks like an attempt to take the attention away from the first (Arnold) and shift it to the second one (Fraser Stoddard), even if the first is brand new and apparently includes an extreme case of falsification of multiple results for a key aspect of the work, whereas in the second one it is “only a picture” (see Blatnoi’s comment above, judgements may differ.)
Falsifications occurring by coworkers may have something to do with the general climate and social interactions in a research group. There might be female specific aspects to this, maybe not; but why so anxious to not allow discussions from coming up?
There are enough standout cases (empathy research, astrophysics, junior biochemistry researchers in Germany) which leaves me wondering if the preferential support for and special attention placed upon young female scientists in recent years may have put them under particular pressure to be successful. Which may sometimes have consequences.
I presume you a male, discussing here a theory whether women are more prone to research fraud?
In any case, I know 3 male professors who retracted own papers, one case is linked in the article. In 2 of 3 cases the perpetrator was male. Cheers.
Yes I am male and I have observed that political pressure to increase the % of female hirings in academia has led to situations where the pool of suitable candidates had a different % m/f than the % hiring m/f.
A first hypothesis would be that if there is preferential hiring from a smaller pool of candidates (females among applicants), those hired preferentially by “affirmative action/positive discrimination” will have a higher chance to fail, if placed in a difficult position.
Another hypothesis might indeed be that different approaches to organizing a research group and interacting with people could be sex related. Sounds like a rather controversial research question in current times, but certainly there should be enough “gender scientists” who really want to do empirical studies to prove or even disprove their position, don’t you think so?
How about this? There are two paths to a faculty position….the classical, and the non-classical pathway. The classical pathway is where the school wants to fill with the best candidate, doesn’t matter what there race/nationality/sex is, its more whether the applicant brings the school grant money or not. These are usually filled by advertised nation-wide searches. In the non-classical pathway, the candidate knows the individual who wants to hire a faculty member…this individual not be as qualified in the classical pathway, but again, someone likes this individual, and may open up some kind of research associate professor position, which will become tenure if they get grant money.
My hypothesis is that by the non-classical pathway, women have the advantage, because they are being hired by old powerful men. I cannot help but to notice in the school I’m in the people who become faculty by the non-classical pathway are usually very, vary physically attractive women, who, while maybe not brilliant, are “smart enough”. And yes I am a man.
Tell you what. It is true about the second pathway, which is actually the real classical academic route.
But: as a man, all you have to do is to toad up to your academic patriarch, laugh at his bad jokes, serve him loyally, carry his bag etc. As a woman you have to sleep with that old git to get same academic opportunities. And even then he might drop you if he finds someone younger and/or sexually more eager.
Still feel women are in advantage here?
Yes, because the attractive women don’t have to sleep with the chairman of the faculty search committee. Just smile a lot, show a little leg, work the hips a little when you walk out the door, and don’t say anything stupid. Because of this, women do have a slight advantage for looking for faculty positions. Especially if you are a hot “diversity candidate.”
I give up. Hope someone else picks up here.
Im glad you finally admit defeat! Again, this is just my opinion. Makes sense to me, but I confess this is off anecdotal observations, and my understanding of evolutionary psychology. And I am a bitter old man too!
You might have a point with your reference to women/minority quota, but it is not the point you tried to make. When male bigwigs try to push through their lovers into faculty jobs, they will play the woman/minority card, after of course having eliminated the applications from qualified outsiders.
The minority card could also apply to male favourites btw, who, may I remind you, don’t have to sleep with their patrons.
To conclude: evolutionary psychology is a bunch of sexist shite.
Again, I cant help but notice that all of the post-docs promoted to RAP (research associate professors) in my academic institution have almost all been very, very attractive women, and the positions were opened up by old powerful men. I doubt if they are boinking, but I guess its possible. I should state for the record that I believe if the woman is quite attractive, she will have an advantage over ugly men like me (!), and no aggressive flirtation is really needed on her part. Why is this? Let me explain with this delightful anecdotal example.
My advisor thinks that a friend of mine, who is very attractive of the latina persuasion, deserves to be faculty. I think she is smart and capable, but certainly not brilliant. She did become RAP for a couple of years, but unfortunately could not get grants and is now back to being a post-doc.
Its been known for quite sometime that attractive people are perceived as more intelligent than the less so, and I really think that plays a influence in my advisors thinking, and other old male professors. But they would never claim that this might affect their thinking, of course.
But if she were to wear a skirt, it certainly wouldn’t hurt her either. Women have this power that men simply don’t posses.
Oh but attractive men have even more success. Paolo Macchiarini, anyone?
That attractiveness, both male and female can impose dominance, is yet another story.
Otherwise, only in Italy do academics flaunt their affairs. Elsewhere they do it more discreetly and you stay clueless.
Also, don’t forget: for every female PhD or postdoc who entered a profitable affair with an older professor, dozens were sexually harassed , then career -destroyed by that same man.
I agree, sexual harassment is wrong, and should not be happening. And yes, probably more women’s careers have been damaged by harassment than may have benefited from positions opening up for them because they were attractive (amongst her other qualities). Completely agree with that. Also, I probably cannot really understand a woman’s perspective here at all as well, as Ive never walked her shoes…
I think the case with the crystals here may be (somewhat) harmless, but I agree with Leonid’s different treatments of Arnold versus Stoddard.
Arnold quickly and publically corrected the record and took responsibility. Now, this may just be savvy PR on her part, but it is also the right thing to do and mimizes damage (to her and to the scientific community).
About the case here, there was a great reluctance do deal with the issue from the journal, and how the process was handled was not very clear. If the pictures of the crystals are real and no fraud occurred (which I think is possible, but not guaranteed), I can understand the annoyance by the authors of having to deal with this, but then again, this is because of their incomplete description of procedure in the original paper.
NMH, I am also male in an academic environment and have experienced my share of carreer struggles. That being said, I have seen enough (small to medium) instances of sexism to realize that the disadvantages that women face (cumulative over a career and averaged over the group), far outweigh a possible hiring advantage in some isolated cases. Also, argueing that women have an unfair advantage because there are too many successful men and not enough successful women in senior positions?
I heard a number of theories how the authors could have added paraquat carefully, drop by drop, slowly and yet quickly (you see colour starting to change already after 2 minutes). Crazy setups, and every theorist is not entirely sure it would work. And yet the authors never mentioned or discretely referenced their contraption once.
Of course, the original simple and straightforward explanation, that the authors mixed up images, that one cannot be proposed anymore, as it would be an accusation of research misconduct in view of the recent correction. Which makes it all very embarrassing.
Pity the authors refused to make a video or to share original image files for forensics.
As for Arnold, she recognised her failure as supervisor and will surely draw the right conclusions. It is also expected that the university will perform a misconduct investigation. The problem here is that star PIs like Arnold are hardly ever present in the lab, especially since the paper was published just when she received her Nobel Prize. But why don’t you visit PubPeer where many other jet-setting bigwigs refuse to do anything at all about the fraudulent papers they published? For example, I had a former Harvard Med School dean admonish me how dare I criticise the refusal of his personal friend C Ronald Kahn to correct his papers.
What about my explanation that the crystals were grown in the vial and that means they would not be disturbed even by turbulent flow if they are stuck to the glass?
It is not the explanation of the authors, right?
That appears to be true. But as someone who deals with a lot of crystals on a regular basis, it’s a very good explanation. I wonder why the authors didn’t use it?
For one they are not that clever, apparently . Also, Science believed them anyway
Ok, I asked and it seems MOF crystals are made in solution. So there.
I digress. Arnold completely failed in checking the most basic data (HPLC’s, Lab Journal entries) of an ongoing project (measurements must have been performed over quite some time if you do this kind of work) and there was evidently no internal check (another person repeating the same experiments). Furthermore the claim of high enantioselectivity with engineered enzymes in a reaction normally reserved to organometallic chemistry is a strong one, so you have to check the data for sure, not just read the manuscript.
On the other hand, the thing with the photoshoped picture would have been easy to slip over, as it only serves illustrative purposes and the basic result (crystals change color by inclusion of a colored molecule) was not surprising. The key result was in the X-ray structure of the inclusion complex which seems to have been o.k. (so far).
In previous discussions of retractions, due to oversights of checking what the coworkers actually did, there has been harsh criticism in pages like the current one or “Retraction Watch” – which I thought was sometimes overly harsh and did not recognize that coworkers are responsible for their actions, too.
In the current case it appears to have been a clearly planned fabrication of results by the coworker (an academically successful female student who might have put herself under a lot of pressure – or maybe was put under a lot of pressure by Arnold? I do not know), who did not get the hoped-for results.
And it seems some people are suddenly much more forgiving … for reasons I find obvious.
I am confused. Are you not convinced by Stoddart’s and Yaghi’s correction?
Of course MOF crystals are made in solution, just like a vast majority of many other crystals for small molecule X-ray. However, this doesn’t mean that they are loose in solution and they can still be stuck very strongly to the glass if they started forming on the surface. This is what happens for some of the samples of the crystals that I see. You have to scratch them off later even when they are still in solution.
As you say, this whole discussion could be moot since that is not what they are claiming, but you have to be careful with a field where you don’t know much. It could make it really easy for your detractors to… detract I guess.
Heh, before writing I showed the case to some chemists who smiled knowingly. Tbf, an expert on MOFs however chose not to reply. Which I understand.
Place these quotes of Focault logically , one will understand .
“Where there is power, there is resistance.”
‘People know what they do; frequently they know why they do what they do; but what they don’t know is what what they do does.”
“Maybe the target nowadays is not to discover what we are but to refuse what we are.”
“The strategic adversary is fascism… the fascism in us all, in our heads and in our everyday behavior, the fascism that causes us to love power, to desire the very thing that dominates and exploits us.”
“Schools serve the same social functions as prisons and mental institutions- to define, classify, control, and regulate people.”
“A critique does not consist in saying that things aren’t good the way they are. It consists in seeing on just what type of assumptions, of familiar notions, of established and unexamined ways of thinking the accepted practices are based… To do criticism is to make harder those acts which are now too easy.”
Pingback: Bullshitter Mauro Ferrari out as ERC President – For Better Science
Pingback: Science misconduct – For Better Science
Pingback: Corona up your NOse – For Better Science
Pingback: The One-Man Papermill – For Better Science
Pingback: Longwei Yin corrections, or what’s the point anyway – For Better Science