Corante

About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: derekb.lowe@gmail.com Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Emolecules
ChemSpider
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
PubChem
Not Voodoo
DailyMed
Druglib
Clinicaltrials.gov

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
Kilomentor
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
ChemBark
Realizations in Biostatistics
Chemjobber
Pharmalot
ChemSpider Blog
Pharmagossip
Med-Chemist
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
SimBioSys
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Business|Bytes|Genes|Molecules
Eye on FDA
Chemical Forums
Depth-First
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa


Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
FuturePundit
Aetiology
Gene Expression (I)
Gene Expression (II)
Sciencebase
Pharyngula
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net


Medical Blogs
DB's Medical Rants
Science-Based Medicine
GruntDoc
Respectful Insolence
Diabetes Mine


Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem


Politics / Current Events
Virginia Postrel
Instapundit
Belmont Club
Mickey Kaus


Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Druggability: A Philosophical Investigation | Main | Another T-Cell Advance Against Cancer »

August 14, 2013

Nanorods? Or Photoshop?

Email This Entry

Posted by Derek

If you haven't seen this, which goes into some very odd images from a paper in the ACS journal Nano Letters, then have a look. One's first impression is that this is a ridiculously crude Photoshop job, but an investigation appears to be underway to see if that's the case. . .

Update: the paper has now been withdrawn. The chemistry blogs get results!

Comments (27) + TrackBacks (0) | Category: The Scientific Literature


COMMENTS

1. captaingraphene on August 14, 2013 7:31 AM writes...

Subtle manipulation of images and data sets (indeed, sometimes not that subtle) is quite more common in academic research than you might think.

In addition, I don't know whats more troubling here: the fact Pease (the PI) claimes ignorance or his apparent threat of legal action towards those who publicize their scepticisms of the study.

"The university is conducting a thorough investigation for research misconduct." It wouldn't surprise me one bit if this entails Pease to 'investigate' himself. There is often either no or barely any oversight for tenured professors.

Permalink to Comment

2. a on August 14, 2013 7:37 AM writes...

Assistant professor = not tenured!

Permalink to Comment

3. will on August 14, 2013 8:02 AM writes...

The sad/scary thing is that it seems like Pease himself threatened the writer with a libel suit for publishing the story. Sounds like SLAPP territory to me.

Of course, I can imagine his anxiety - assuming the data was fabricated, whether with the PI's knowledge or not, it's probably a hard thing to overcome in tenure review. Keeping the story off the internet would make it easier to slide into another position somewhere else

Permalink to Comment

4. Anonymous on August 14, 2013 8:10 AM writes...

At least if his academic career is finished he could still get a job as a graphic designer. Not.

Permalink to Comment

5. Anonymous on August 14, 2013 8:16 AM writes...

ā€¯Fabrication of Highly Uniform Nanoparticlesā€¯?

Sure looks like they were "fabricated"! :-)

Permalink to Comment

6. captaingraphene on August 14, 2013 8:41 AM writes...

@2

You are right. My mistake. Anyhow, it remains strange how he, in his non-tenured capacity, can threaten legal action on behalf of the University of Utah.

Permalink to Comment

7. Validated Target on August 14, 2013 9:08 AM writes...

Perhaps some knowledgeable person can explain some details of those images. For example, I THINK that the images show (1) smooth grey areas = TEM grid = holey carbon (2) very light areas = wide open spaces (holes) in the grid (3) very dark small pairs of objects = nanorods. Usually, the holes should be empty. Long objects (nanotubes) that span the hole and are supported by solid grid material on both sides can appear in the very light areas. Small objects should just fall through the holes or not adhere (there is nothing to adhere TO!) during sample prep. That is, objects floating in the holes was the first clue? IS THAT ABOUT RIGHT?

The TEMs of Fig 4 do look kind of splotchy (light rectangles along the bottom).

What else betrays these as fake images?

Permalink to Comment

8. Wile E. Coyote, Genius on August 14, 2013 9:30 AM writes...

@Validated Target:

I don't have access to the paper as it is behind a paywall, but the abstract indicates that they sprayed these onto a matrix. The background in the image is the matrix - and that matrix may not be uniform - I don't know what it is. I would not be surprised to see a matrix that is not perfectly uniform. I believe you are misinterpreting the images. The smooth grey areas are not TEM grid. The light gray areas are not spaces between the grid. TEM grid is normally a wire grid that is completely black (in an electron micrograph - completely electron dense - electrons don't pass through it) and is not in the field of view of any of the images here. Everything that you are seeing is between the wires of the grid. When doing TEM, at least for cells or tissues, the material is embedded in plastic and sectioned with an ultramicrotome. These sections are then placed on grids for EM examination. It may be that the matrix that these are sprayed on is already on the grid or is placed on the grid. Therefore, they won't fall through the spaces on the grid, and the grid is never seen since the grid is of no interest. In fact, it is "bad form" to have the grid included in your electron micrographs, so fields of interest are shown that do not include the grid. At these magnifications, the grid wires would look to be "miles" apart.

Are the images doctored? Probably, but not in the way that validated target is proposing.

Permalink to Comment

9. anonymous on August 14, 2013 9:38 AM writes...

In the image on Chemjobber's site, 1) the duplication of two rods in the light area (there are dark spots floating in space in the same spots near a pair of nanorods) 2) the sharp edges around the nanorods in the light area (the area immediately around the rods doesn't match the background and the boxes containing the different backgrounds have rather sharp edges which aren't consistent with any other features in the rods and would be extraordinary in themselves) and 3) the places where two rods in the light area overlap don't show overlap, but instead show sharp edges and lighter backgrounds from the boxes containing the rods, which doesn't fit physical object behavior but does fit cut-and-paste image manipulation. I imagine there are more, but I haven't looked much.

I wonder a) who looked at these (from advisor/supervisor to reviewer to editor) - the images are some of the key work of the paper and the inconsistencies are pretty blatant so that someone ought to have noticed them, and b) why anyone with a functioning brain would insist on threatening libel when the evidence for manipulation is publically available and is almost certain to be revealed somewhere (not only don't you keep it quiet, but you make sure that when it is found, people will accord you a reputation you probably deserve and will dig and dig until what remains of your reputation is in shambles). Even if the information isn't publically available, people don't like being threatened, and they will start digging into what you didn't want them to dig.

Permalink to Comment

10. Bauke on August 14, 2013 10:16 AM writes...

@5: just had the same thought ;-)

Permalink to Comment

11. Flatland on August 14, 2013 11:48 AM writes...

As brought up in the link, Anumolu is listed as both the first author and the corresponding author, and is using a non-institutional email address.

Seems that this might be a case of someone graduating and being unable to get a job, and deciding that the CV needs some added 'heft'.

Permalink to Comment

12. Rcyran on August 14, 2013 11:55 AM writes...

No worries about libel. Under US law, a person can only be convicted of libel if they knowingly publish a falsehood. Key word is knowingly.

There's no better tell in journalism that a story needs to be published than a threat of a lawsuit!

Permalink to Comment

13. chemist on August 14, 2013 12:34 PM writes...

11 @ Flatland, Did you dream about it? Pease is the corresponding author and his institutional email is listed in the paper and supporting info...Or you are so much prejudiced against students?

Permalink to Comment

14. CaptainObvious on August 14, 2013 12:52 PM writes...

I work close enough to the field to be familiar with this. This is not even subtle, it is a photoshop job. The only half reasonable explanation if the TEMs of the rods were taken at a higher resolution and duplicated over a coarser TEM.
The biggest mystery is how its possible for this to get past the review process...

Permalink to Comment

15. ClutchChemist on August 14, 2013 1:45 PM writes...

I believe I read on another comments section the suggestion that the journal editors published the paper just to make the fradulent figures public. Otherwise, the paper gets rejected and the prof/students can fix the images, submit elsewhere, and no one gets caught. Is this in the realm of possibility?

Permalink to Comment

16. The Iron Chemist on August 14, 2013 1:50 PM writes...

@13: Chemist, not a dream as I was the one who raised the point on the Chemistry Blog and as far as I can tell, am not a figment of anyone's imagination. The student has the asterisk behind his name identifying him as the corresponding author and a gmail account listed in the contact information. You may be thinking of the second paper mentioned on the blog; all Pease there.

The corresponding author issue doesn't necessarily absolve Pease of blame, but it's strange enough to merit a little more scrutiny before demanding the end of the man's career.

Permalink to Comment

17. anon on August 14, 2013 3:10 PM writes...

1) I would assume after the kerfuffle over Mirkin's publication sent in by a postdoc, that journals would make sure that all the authors knew about the publication and reviewed it. It's possible they didn't, but that is a high barrier of incompetence to surmount. Maybe faked figures could have been substituted during revision and not caught, but the only reason that would make sense would be if the grad student had a career death wish, which is possible, but seems unlikely.

2) Pease's threats/enjoinments not to publish information on the issues of the paper don't speak well of him or his role in them. He could have referred the matter to Utah, or simply said that he couldn't comment while it was under review by Utah, but (according to ChemBlog) he didn't. Again, maybe he was overtaken by the whole thing, and responded irrationally because he didn't expect falsification to happen to him (and he had no role in it), but the nature of the response is likely to make people think that he had something to hide.

Fortunately, everyone's career is not in the hands of blog commenters. On the other hand, considering Columbia's machinations over Sames/Sezen (and the silence of nearly every institution or person involved in it to any degree), it is difficult to trust that people will know enough to make informed decisions about the people, research, or processes in question if they wait for the parties involved to tell them.

Permalink to Comment

18. anon on August 14, 2013 4:09 PM writes...

Why doesn't the Journal investigate? Instead they leave it to the University of Utah? Sounds like a conflict of interest, or am I mistaken?

Permalink to Comment

19. Renee on August 14, 2013 8:10 PM writes...

Makes you wonder how carefully (or not) the reviewers viewed the TEM images.

The only people who benefit from this Nano Craze are the makers of electron microscopes. Science certainly isn't benefiting from it.

Permalink to Comment

20. Justin Peukon on August 15, 2013 7:29 AM writes...

Not a "Photoshop" job. Rather a "Paint" collage.

Permalink to Comment

21. samadamsthedog on August 15, 2013 12:08 PM writes...

@Renee 19.

Reviewers do not usually take a forensic stance when commenting on a paper. They would not usually look at images like this and ask whether they fraudulent; maybe if they stopped to ask the question they would have thought them so, or maybe not. They would be more likely to assume the images (and other data) are real and ask, on that assumption, whether the authors' conclusions are justified.

In a previous thread, I commented that, as a reviewer, I don't look at the supplementary material unless I have to. I assume that the material has been reported honestly in the article, and only if I don't understand something in the article do I go to the supplementary material to attempt to understand it. If I do need to go to the supplementary material, I usually recommend that the exposition in the article should be more complete, so that the reader does not have to consult the supplementary material.

On the other hand, if, as a reviewer, I were asked to do forensics, then I would have to look at the supplementary material with a jaundiced eye and even request and examine 10x the amount submitted, or than is generally required to fulfill the need that supplementary material is supposed to provide.

But if, as a reviewer of a paper for a journal, I was in fact asked to do forensics, I would decline on the grounds that (a) I have never done this before and am not sure I am qualified; and (b) it would take far more time to do it well than I can spend.

It's true that some things might just jump out at me as unlikely, but a lot more would go past unless I were focusing on uncovering fraud, rather than focusing on whether the data, assumed to be honest, justifies the conclusions.

Permalink to Comment

22. Anonymous on August 15, 2013 1:13 PM writes...

Journals should insist that all authors sign a statement to the effect that the results reported are true and real to the best of their knowledge, or accept that they may have to pay a fine to the journal for damages, if results are found to be knowingly falsified like this.

Permalink to Comment

23. anon on August 15, 2013 1:50 PM writes...

It doesn't seem like it should be a reviewer's job to investigate possible fraud, but to make sure that the data justify the conclusions (does the data have enough power to justify the author's conclusions from it?). However, this picture doesn't look good at all, and people have commented that at least one of the pictures in the main document (Figure 4) had similar sketchiness, and people presumably would expect a reviewer to notice such things. In addition, the images are the key data in the paper, and probably merit scrutiny, even if not in the body of the paper.

The "make up the EA" article, on the other hand didn't depend on the EA to power its conclusions, though browsing through the SI probably would have turned up the statement and stopped the review process in short order. In that case, it seems embarrassing, but not a blatant error (if one at all - they may not have been able to see it) for a reviewer.

When SI was not included with a large proportion of documents and was difficultly available to readers when it did exist, it makes little sense to review the SI in depth (though it made little sense to ignore it if present). However, when many documents rely on the SI to show data that journals (or the authors, due to page charges) do not want to include in the main text, and when it can be assumed to be readily available to readers and likely to be seem by (some of) them, it ought to be reviewed as a significant part of a document.

Permalink to Comment

24. Florian on August 16, 2013 3:55 AM writes...

By definition, scientists seek the truth.

So fraud - even if it is for proving facts that you are sure are there but you cannot prove - is one of the worst things you can do.

When I am reviewing a paper (I do that about 10-20 times per year), I am not really focussing on whether the data is real or fake. My subsection of science is small enough that I know most of the scientist there personally and I know whose data is really trustworthy. Most of the flaws in the data are artifacts, which were not properly compensated.

However, when looking at the posted images, it seems not understandable to me, how the refees have not picked that up. When looking at Fig. 4 from the paper in question, I can clearly see some strange halos at page width zoom on a 15"-Laptop. This should ring an alarm bell. The least I would have done as a referee is to write as a comment to the editor that this figure seems fishy and that he should have a closer look at it too.

However, in my experience, reviewing isn't always done the way it should, but that is mostly restricted to low impact publications.

Permalink to Comment

25. Anonymous BMS Researcher on August 17, 2013 11:10 PM writes...

Back in the early 1990s, before most journals established policies on what was and wasn't acceptable with digital images, a lot of people did what I thought was excessive cleanup -- fiddling with balance and contrast to make bands on gels and FISH mapping look cleaner than they truly were. But there's a big difference between cleaning up noise versus adding signal where none exists. I sometimes saw what I felt was questionable procssing -- and sometimes said as much to people at the time -- but I sure never witnessed out-and-out fraud.

Permalink to Comment

26. Validated Target on August 18, 2013 8:22 PM writes...

@7 and 8: I took another look on a better computer and blew up the images. Amazingly obvious to me now. Different backgrounds; sharp edges; identical replicates; etc.. I can almost smell the glue wafting from the computer from that cut and paste job. I hope somebody gets nailed.

Permalink to Comment

27. Vader on October 23, 2013 9:22 AM writes...

"No worries about libel. Under US law, a person can only be convicted of libel if they knowingly publish a falsehood. Key word is knowingly."

While Utah has a criminal defamation statute, it's something of an anachronism, and it's very unlikely to be enforced even if the mens rea requirement can be satisfied. So "conviction" is not quite the right word here.

The more realistic concern is of a civil tort. For this, no criminal intent is required. That is, you can sue someone for libel in almost any U.S. jurisdiction purely on the basis that what they said was false and defamatory. You don't have to prove that the defendant knew it was false.

Unless the suit falls under the Sullivan rule, applicable if the plaintiff is a public official or public figure. Such plaintiffs are required to show actual malice, meaning that the defendant knew or should have known the published defamatory claims were false, or showed reckless disregard for whether they were false.

It would be interesting to see whether the court regards an obscure, untenured university professor as a public figure by this standard. I think they should, and suspect they probably would, but a lot of courts have produced a lot of rulings in the past that went against what I thought they should and would do.

Permalink to Comment

POST A COMMENT




Remember Me?



EMAIL THIS ENTRY TO A FRIEND

Email this entry to:

Your email address:

Message (optional):




RELATED ENTRIES
One and Done
The Latest Protein-Protein Compounds
Professor Fukuyama's Solvent Peaks
Novartis Gets Out of RNAi
Total Synthesis in Flow
Sweet Reason Lands On Its Face
More on the Science Chemogenomic Signatures Paper
Biology Maybe Right, Chemistry Ridiculously Wrong