When rigour deserts science we are left with quackery

This is a guest post by Dave Fernig, Professor of Structural and Chemical Biology at the University of Liverpool

Standards? Rules? Norms? Guidelines? What for? These days in the world of science I can only find outrageous manipulation, falsified data, stunning self-justifications and serial, unparalleled fraudsters. We need to enforce protocols, we need to reinforce canons! (Even if only for shooting at those fraud-riddled crooks.)

Examples abound in Retraction Watch, my favourite news outlet. By the way, you should all get in the retractionwatching bandwagon! It’s real fun (and free!) to condemn potential criminals with callous comments. Gives you energy to crank through the day!

For example, the stem-cell paper published in Cell was full of fake images. Cell pulled the “Dictionary of Euphemisms” off the cuff to make weak excuses: “we have no reason to doubt the thoroughness or rigor of the review process“. An avid retractionwatcher would immediately tell that what Cell claims is that they do not doubt the thoroughness OR the rigour of the review process. So which one do you doubt, you rogue Cell makers! Euphemisms, big time! That’s how you sell your soul to Sauron? Chaps, this would have been more novel: ‘The authors were quacks, the editor rammed the paper through peer review, the reviewers were scarecrows, and we can’t wait for the fountain of citations to flow’.

Another sickening case is that of Maya Saleh. It is worth quoting the conclusions from an investigation by McGill: “two figures in [a] Nature paper had been “intentionally contrived and falsified.” One of those figures was duplicated in a PNAS paper, which also contained an image that had incorrectly labeled some proteins.

What happens? Nature issues corrections, as the committee recommended!

Amazing, really. The other day in the classroom an excellent student of mine cheated, but I didn’t notice (I was busy recalling the names of all the dwarves at Frodo’s dinner). But whistle-blower Rapha discovered the cheating (this chap is a goldmine!). We then formally investigated by asking the student for the raw thoughts, and immediately corroborated that the student had indeed cheated. Did we allow the student to “correct” the work outside the exam and at their leisure??? Yes, precious, she could. And then we takes it once they’re dead.

Of course not, the student gets a zero! This brings me to Francesco Stellacci. You will all recall that there have been a gazillion of thoughtful concerns raised about Stellacci’s work. It’s all fake (I mean, the work). The first proof was aired with the first post on the popular blog of my fellow Rapha when he finally published his article in Small. The chap had to suffer for years a litany of rejections and bigoted reviewer reports from more unscrupulous journals. To be fair, however, those unnamable venues were not small for champ Rapha!

But since Rapha surfaced victorious from the great battle, lots of elfs folks (myself, Rapha, Professor Moriary and uncountable others) have enumerated too many issues to be counted: image re-use, using again an image from a sample that was also used to describe a completely different experiment, the publication of a published image in another journal without attribution, the usage of an old image in a new paper, the utilisation of fake images from previous work, the misuse of published images, and the inclusion of images that were not new. And we should not forget the stonewalling efforts of a legion of third parties (Philip M., P. Moriarty, Professor Moriarty, Holmes’ nemesis,…) to access raw data. We wants it, we needs it. Must have the precious. Sneaky little FS. Wicked, tricksy, false!

Still, some piffling progress has been made; two corrections in infamous journals and the appearance of a handful (but only a few!) of fake raw images. I have had a look at these. I know I should be examining the original files through Gimlion (son of Glóin), but from the little SPM work I re-used from a colleague some 15 years ago I know that if something is not obvious in the TIFF, it isn’t there.

And obviously there are no stripes on the raw images of Stellacci. I only see noise. How then to get the plain vanilla images published in a whole series of papers since 2004? Elementary, my dear Watson: manipulate the images. Matey Moriarty knows this well (caution: the chap is a bit prolix).

Now to this blog. As I was reading previous posts, I realised that I was getting a first hand insight into the filthy mind of a science fraudster. I went to re-read the interview of Stapel, the fraudulent psychologist. The vanity, pomposity, conceit, treachery, excuses and apologia of this foul sounded familiar… Then I got it: for the lord of Mordor, Stellacci!

He wrote “But hey, how did the guy make that glamorous image for Nature Materials? I can’t get it right from the raw data! See, I told you guys it was a fake.”

This fellow is completely caught up in his folly logic: he concludes that if he can’t get the published images from his raw data, then the published images are fake. Imbecile! We all know that experiments are messy! That he can’t get those published images right from his fake raw data (not surprisingly, he is no Einstein) doesn’t mean that someone else can’t. My smarty blokes Rapha and Moriarty surely can!

Without doubt, he is a purely simple fraud. I have informed his former employer, MIT, and his current one, EPFL. As giants are only so big and dragons can only fly so far (but I got my precious, ombudsmen!), in due curse I shall retractionwatch Stellaci’s papers or loathe a truckload of corrections and the fattening of the “Thesaurus of Euphemisms“.

Are we wasting out time bringing these problems to the fore? NO, I don’t waste my time because:
0. I am paid from public money to retractionwatch and to give zeros to cheating students.
1. Fellow Rapha and I don’t publish non-reproducible science in journals (we only post it on our blogs).
2. Science fraud neither kills giants nor dragons! (Mustn’t go that way! Mustn’t hurt the precious!)

UPDATE: As it is clear from the text above, attributing the authorship of this blog to Stellacci was beyond doubt a guess or a hypothesis (guess which!). Now, fellow Rapha has told me that this is his candid blog; seems reasonable, as he could do with the attention of more trolls and orcs. So my guess was wrong, and I retract it (Stellacci, do you hear this, my fellow? I am not afraid of retractions; I eat them for breakfast!)

Data re-use update

Every time I look at a stripy nanoparticle paper, I see the same images. I am not at all surprised about that, because hey, once you’ve convinced everyone that your nanoparticles got stripes by tricking a glamour magazine into publishing artefacts, you’ve got no need to spend time fabricating more images with stripy particles; you can live the dolce vita by analyzing and re-analyzing the same artefacts. Make no mistake, F. S. has been preaching to the converted.

But I am not that credulous; I don’t buy the stripy religion. Je sers la science et c’est ma joie! (I save science and this is my job!). Hey, I publish in a blog (well, now two of them) for anyone to see, not in those glamour journals that feed sexy stuff to the ingenuous.

I have to admit, however, that the first time I read the stripy paper published in the Journal of Scanning Probe Microscopy, F. S. tricked me into believing that there was new data. Hey, it even occurred to me that F. S. is not such a wicked folk after all!

Oh, boy, was I wrong (for all you out there, non-believers: see, I admit I may be wrong once in a long while!). I have now revisited that paper, and in fact ALL of the data in it have been re-used!

Wait, there is more. They cite their previous work, but go to great lengths to avoid writing that the data are re-used:

Much of the challenge lies in the sample preparation and in the image interpretation.43,44

[…] (see Experimental Details section).44,45

The synthesis procedures used for the particles described in this paper are derived from previous reports56–58,68 and have been discussed in detail in previous reports from this group.43–45,70

You have to go to the end of the paper to find in small type where the figures first appeared: Ref. 43. is Nat. Mater. 3, 330 (2004); Ref. 44. is J. Am. Chem. Soc. 128, 11135 (2006); Ref. 45. is J. Phys. Chem. C 112, 6279 (2008).

How cheeky. The figure legends do not indicate clearly where the data analyzed come from! Those smart-asses at EPFL made me work hard. I had to read the text, and then figure out that those small numbers meant that relevant data had been published before, go to the end of the paper, look for the numbers in a long list, go search for the papers referenced, and compare the data. Only to find out that the data is the same! RE-USED!

F**k man, what sort of trick is this? If you re-use a figure, write in the figure caption ‘Data self-plagiarized from previous work published in glamour journal X’.

But no, they deliberately hid the re-use of fake data. This shall not be allowed. Hey, they can’t publish the same fake data more than once and get away with it!

No doubt the paper should be retracted. I’ve gotta talk to my mate Dave Fernig; he knows how to set those journals straight.