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

« Day Off! | Main | GE Healthcare's Idiotic Libel Suit »

December 21, 2009

Faking X-Ray Structures. . .For Fun? Or Profit? Or What?

Email This Entry

Posted by Derek

Well, this isn't good: an ex-researcher at the University of Alabama-Birmingham has been accused of faking several X-ray structures of useful proteins - dengue virus protease, taq polymerase, complement proteins from immunology, etc. There have been questions surrounding H. M. Krishna Murthy's work for at least a couple of years now (here's the reply to that one). The university, after an investigation, has decided that 11 of the published structures seem to have been falsified in some way and has asked that the papers be retracted and the structures removed from the Protein Data Bank.

The first controversy with these structures was, I think, the one deposited in the PDB as 2hr0. Here's a good roundup of what's wrong with it, for those of you into X-ray crystallography. And as that post makes clear, there were also signs that some other structures from this source had been suspiciously cleaned up a bit.

So how do you go about faking an X-ray, anyway? Here's some detail - basically, you could take something that's structurally related (from a protein standpoint) but crystallographically distinct, and use that as a starting point. As that post says, add some water and some noise, and "bingo". The official statement from UAB's investigation gives you the likely recipes for all eleven faked-up structures.

As for Dr. Murthy, he left UAB earlier this year, according to this article, and the university says that they have no current contact information for him. If these accusations are true, he's spent nearly ten years generating spurious analytical data. What, then, do you do with that skill set?

Comments (30) + TrackBacks (0) | Category: Analytical Chemistry | The Dark Side


COMMENTS

1. Vader on December 21, 2009 10:04 AM writes...

"If these accusations are true, he's spent nearly ten years generating spurious analytical data. What, then, do you do with that skill set?"

East Anglia CRU?

Discovery Institute?

Just to cover the spectrum.

Permalink to Comment

2. Pinko Punko on December 21, 2009 10:12 AM writes...

Nice try, Vader.

Permalink to Comment

3. gyges on December 21, 2009 10:49 AM writes...

Why aren't these people prosecuted?

Permalink to Comment

4. jjt on December 21, 2009 10:54 AM writes...

Lobbyist?

Permalink to Comment

5. Hap on December 21, 2009 11:16 AM writes...

Political think-tanks are good places to start. (There's always spots to help people with preconceived dishonest positions to plaster over their dishonesty with a scientific veneer and some bad but impemetrable math). I have to figure that there are some Ponzi schemes out there looking for helpers. Of course, as long as he can get funding, I would have figured some university will pick him up, saying that "people are being too judgemental" and "we're giving him a second chance". So long as the overhead checks keep coming, they'll be willing to help. Does Cordova's uni have an opening?

Permalink to Comment

6. Sili on December 21, 2009 11:30 AM writes...

Sorry, Vader, but the Disco 'tute has never put that much effort into anything that closely to science. (But I do still have to thank them for that Beware the Believers video - pure gold.)

Can't have been for fun. From what P2111 sez this was done much too ineptly to have been a labour of love*. 30 Å empty spaces? Why did no one see this earlier?

*Of course, I'm hardly representative in that I have more than once spent an afternoon trying to model disordered counterions and solvents (on ShelX**, where one has to cheat), only to discard it in the end, because leaving in ridiculous anisotropic vibration parameters gave the best fit anyway.

**Speaking of scientists that look like actors and vice versa: I've only seen Sheldrick once, but I was rightly warned that I would not be able not to mentally equip him with a big, white, Burmese cat.

Permalink to Comment

7. RB Woodweird on December 21, 2009 11:37 AM writes...

Dr. Murthy will obviously be found in the near future behind a big desk at Merrill Lynch. He will be in charge of your 401K plan and will be making $500,000 per year before bonuses.

Permalink to Comment

8. dearieme on December 21, 2009 1:28 PM writes...

"What, then, do you do with that skill set?" Search for weapons of mass destruction?

Permalink to Comment

9. Chemjobber on December 21, 2009 2:38 PM writes...

Appraise homes in Southern California?

Permalink to Comment

10. Martin on December 21, 2009 6:21 PM writes...

One thing I've been pondering in all of this is what is the $ cost of all of this? How many dollars of Industry and government grant money has been wasted following up this spurious structural biology? From personal experience we were a bit suss on the 1BEF/1DF9 Dengue protease structures from when they came out, but as they were all there was we used them in some structure-based Drug design work (most of which we completely had to redo when the later Novartis structures came out). Now academic groups like ours are hardly likely to go and sue this guy for lost money but there is a real cost here, and not just to the reputation of science.

As someone who has only ever contracted for industry, and never been "inside", I can only presume that industry always treats stuff like this with a fair degree of scepticism and doesn't invest too much on the back of it, otherwise we'd be seeing lawsuits wouldn't we?

The other point I'd like to make is that all these papers made it through review, and that the review process clearly didn't pick it up. Expert readers out in the field who shouted "enough!" did. This has lead to calls to make raw data available for review, but in reality, what reviewer for JBC/Science/Nature wants to download gigabytes of data to verify a structure?

Permalink to Comment

11. Nat on December 21, 2009 7:06 PM writes...

Martin: I suspect the damage caused by these structures is less than that of the five retracted papers from Geoff Chang's lab a few years ago, in which the data was genuine but grossly misinterpreted. At any rate, this is one reason why the NIH has an Office of Research Integrity. People have gone to jail for defrauding the NIH in the past.

As someone who has only ever contracted for industry, and never been "inside", I can only presume that industry always treats stuff like this with a fair degree of scepticism and doesn't invest too much on the back of it, otherwise we'd be seeing lawsuits wouldn't we?

Actually, I'm pretty sure I read somewhere that the dengue virus protease structure was being used for drug design. I'd be surprised if industry wasn't using any available academic structure of a potential drug target for design purposes. Unless the university files a patent, there's nothing to stop any interested company from downloading the coordinates from the PDB. I'm not a lawyer but I doubt this would give them grounds for a lawsuit, unless they paid UAB for commercial rights.

This has lead to calls to make raw data available for review, but in reality, what reviewer for JBC/Science/Nature wants to download gigabytes of data to verify a structure?

For reviewers, a comprehensive validation report should be sufficient. This wouldn't stop a more competently executed fraud at the peer review stage, but it would certainly make what Murthy did impossible in the future.

Permalink to Comment

12. milkshake on December 21, 2009 7:10 PM writes...

We had a gentleman in charge of animal group in our institute, and after a long period of problems with their assays they finally started producing some encouraging results, but the guy left (for a top job at Merck).

He never showed the raw data to anyone, they were located on a biology server inaccessible to chemists. Only after this guy left someone checked the raw data stored in his computer - and it turned out that everything that he presented on the project meeting was "cleaned" by hand to produce the desired results... Apparently they are re-doing everything now at the institute because they urgently need the data for publications and grant applications - and they don't know how far back this data massage goes/

Permalink to Comment

13. Nick on December 21, 2009 8:07 PM writes...

And another! These guys faked data to get published in Acta Cryst C and Acta Cryst E!

http://journals.iucr.org/e/issues/2010/01/00/me0406/index.html

Permalink to Comment

14. Anonymous on December 21, 2009 8:19 PM writes...

Q: What, then, do you do with that skill set?"

A: Head of voter registration in Chicago (or any other major city for that matter)? Probably highly quailfied for a position at ACORN as well...

Permalink to Comment

15. Sili on December 22, 2009 8:40 AM writes...

These guys faked data to get published in Acta Cryst C and Acta Cryst E!
WHYYYYYY?!

--o--

Sheesh, Acorn?! I can understand why you wouldn't wanna put your name on that crap.

Permalink to Comment

16. Morten G on December 22, 2009 9:01 AM writes...

@Sili: I was just baffled that G-Shel is actually able to read a computer screen. I know scientists take pride in not being vain (see, first law of thermodynamics even applies to deadly sins) but really... snip snip.

Permalink to Comment

17. HappyDog on December 22, 2009 9:11 AM writes...

Actually Mat, the approach usually taken in industry is to take the publicly available xtal structures with a grain of salt. Careful modelers read the header files and pay attention to the B-factros and occupancy factors of all the atoms. Industry crystallographers usually attempt to get a crystal structure in-house if for no other reason than to confirm what was published in the literature. If our crystallographers can't repeat the results, then there is a problem.

I was actually a grad student in a different department at UAB while Krishna was there . It could be that Krishna was really just completely incompotent and he felt pressure to get results for DeLucas' multi-million dollar NASA grant, so he started faking structures. I'd say faking data, but apparently he didn't even bother to do that with all of his reported structures since he never even deposited structure factors for several pdb files. Which begs the question - shouldn't that raise red flags if a crystallographer won't even deposit the SF's?

Permalink to Comment

18. Nat on December 22, 2009 10:28 AM writes...

Industry crystallographers usually attempt to get a crystal structure in-house if for no other reason than to confirm what was published in the literature.

Fair enough - although I assume they'd still end up using the original structure for molecular replacement. Based on what I've seen about the original fake Dengue protease structure, this probably would have been impossible.

I'd say faking data, but apparently he didn't even bother to do that with all of his reported structures since he never even deposited structure factors for several pdb files. Which begs the question - shouldn't that raise red flags if a crystallographer won't even deposit the SF's?

It seems bizarre in retrospect, but two decades ago the idea that deposition of coordinates should be mandatory met with fierce resistance from some crystallographers. Structure factor deposition was probably equally controversial (1998, maybe?) - I've read some of the opinions of prominent scientists who objected to the proposal and they're almost embarrassingly narrow-minded and elitist. When I started grad school (2003), the majority of journals still didn't require data deposition, and it's only in the last few years that this has changed. Fortunately, you'll have a hard time finding crystallographers who are still willing to publicly defend withholding data. Disasters like this should reinforce the trend towards greater openness.

For what it's worth, it looks like Murthy started faking data even before he arrived at UAB.

Permalink to Comment

19. HappyDog on December 22, 2009 11:40 AM writes...

"Fair enough - although I assume they'd still end up using the original structure for molecular replacement. Based on what I've seen about the original fake Dengue protease structure, this probably would have been impossible."

That's usually the plan. If you use MR and get a large difference in the calculated R and free-R values, something is fishy. This way, MR at least gives you a quick guide as to whether or not the published structure is reasonable. Industry crystallography labs usually seemed to be too overworked to completely solve every new structure de novo.

Permalink to Comment

20. Sili on December 22, 2009 11:53 AM writes...

Morten,

I think we may may be focusing on different things.

Permalink to Comment

21. MAD on December 22, 2009 12:31 PM writes...

I think the fact that nobody noticed for so long showes that whatever he was working on wasnt interesting enogh for anyone to bother to even repeat and build something practical off of.

Permalink to Comment

22. J-Bone on December 22, 2009 1:14 PM writes...

"The other point I'd like to make is that all these papers made it through review, and that the review process clearly didn't pick it up."

More proof that the peer review process is completely flawed. I understand that some professors do good work and a journal will be more apt to accept their work on that basis, but should K.C. Nicolaou really be allowed to publish his Photobucket collection in JOC? (2009, 74, 951-972)

Permalink to Comment

23. cookingwithsolvents on December 22, 2009 1:25 PM writes...

@ 21

or it takes that long for someone to express, purify, and crystallize a few mutants and then they find out the xray doesn't look anything like the published structure.

ugh what a mess. . .

Permalink to Comment

24. Morten G on December 23, 2009 7:15 AM writes...

Thanks Nick! That is amazing and amazingly weird. How did publications get more important than science?

Permalink to Comment

25. Old Timey on December 23, 2009 1:08 PM writes...

"What, then, do you do with that skill set?"

He can go work for Al Gore.

Permalink to Comment

26. Sili on December 23, 2009 1:20 PM writes...

Yeah, right.

Permalink to Comment

27. Jason on December 23, 2009 10:47 PM writes...

Perhaps he can join forces with Mohamed El Naschie, about whom you have written on before, and found a new journal.

Publish Your Work The Easy Way

Permalink to Comment

28. dvrvm on December 24, 2009 5:37 AM writes...

Xenobu Research Institute, obv.

Permalink to Comment

29. KML on December 24, 2009 8:01 AM writes...

My PhD boss used to tell us that ´Publications are consequences of research and not the goals´...many of us are really forgetting it or we are in the wrong field..

Permalink to Comment

30. Pats on December 24, 2009 11:07 AM writes...

The paper on 'oxidation of alcohol by NaH' has been withdrawn today...

"Reductive and Transition-Metal-Free: Oxidation of
Secondary Alcohols by Sodium Hydride [J. Am.
Chem. Soc. DOI: 10.1021/ja904224y]. Xinbo Wang,
Bo Zhang, and David Zhigang Wang*
This manuscript has been withdrawn for scientific reasons."

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
The Worst Seminar
Conference in Basel
Messed-Up Clinical Studies: A First-Hand Report
Pharma and Ebola
Lilly Steps In for AstraZeneca's Secretase Inhibitor
Update on Alnylam (And the Direction of Things to Come)
There Must Have Been Multiple Chances to Catch This
Weirdly, Tramadol Is Not a Natural Product After All