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

« Reality-Based Biotech Investing | Main | Ancient Modeling »

April 3, 2014

More Fukuyama Corrections

Email This Entry

Posted by Derek

The Fukuyama group has another series of corrections out, this time in JACS. Here's one, and the other follow right behind it in the ASAP queue. This adds to the string of them in Organic Letters. It's more whiteout stuff - vanishing solvent peaks and impurities. These presumably don't affect the conclusions of the paper, but they don't make a person any more confident, either. One hopes that these high-profile cases will shake people up. . .

Comments (24) + TrackBacks (0) | Category: The Scientific Literature


COMMENTS

1. See Arr Oh on April 3, 2014 12:16 PM writes...

Hey, and one in ACIEE! Nearly all have the same co-author, FYI: http://justlikecooking.blogspot.com/2014/04/fukuyama-round-2.html

Permalink to Comment

2. TGS on April 3, 2014 12:36 PM writes...

Note: in the current issue of Angewandte Chemie there is a corrigendum for another Fukuyama paper!

Permalink to Comment

3. anchor on April 3, 2014 12:42 PM writes...

May be he is having his "Epiphany" moment!

Permalink to Comment

4. Anonymous on April 3, 2014 1:12 PM writes...

i hope this does not lead to the usual 'pile-on' we see in this blog. The blog seems a little skewed toward 'be careful about trusting work from academia'. Probably as much fudging in industry, just never is exposed to much peer review.

Permalink to Comment

5. Hap on April 3, 2014 2:00 PM writes...

I would doubt that - 1) yields aren't as important in industry (discovery has "enough" and "not enough", and fudging published yields for scale-up chemistry won't change the economics, so that the accountants will probably notice and then so will the managers) and 2) because synthetic sequences aren't as long, it becomes more likely that someone will do your reaction again, notice the discrepancy, and nail you. Also, working in groups probably helps lower the chance that you'll lie about yields, because someone will probably notice.

I don't know that this dishonesty is particularly academic in nature, but rather it's academic because no one else does total synthesis. Since there isn't a practical reason to make a natural product (in most cases), but artistic ones, people come up with arbitrary constraints for syntheses to distinguish good ones from bad ones (first to product, yield, atom economy, step economy, etc.) to attempt to remove judgment from the valuation. It's unlikely that anyone will ever repeat a synthesis (or its individual steps), and because the overall yields are important for the status of publications, grants, advisors, and students, there's lots of pressure on the synthesizers. Without the counterpressure of others doing your reactions again, there is lots of reasons people might "fix" their yields.

There's pressure on methodology people as well, but the likelihood that someone will use your reactions and verify that they don't work as you said probably decreases the chance of malfeasance.

Permalink to Comment

6. Doug Steinman on April 3, 2014 2:24 PM writes...

Professor Fukuyama was a consultant for the pharma company I worked for. He was by far the best consultant with whom I ever interacted. He was down-to-earth, honest (if he did not Know the answer, which was extremely rare, he would say so) and incredibly helpful. If it truly is the same co-author on all of the papers in question I would suggest that there is where most of the blame resides for this.

Permalink to Comment

7. featherson on April 3, 2014 3:03 PM writes...

@Hap

I totally agree with you and hence not with 4. Anonymous. As a process chemist I know that whatever process I develop will be implemented in the future by someone be it manufacturing or a CMO. The process has to work. Not the case with academic research. We had a bad experience with a former company I worked for where management believed all of the results from the academites in the company, all of which the process chemist pragmatists scoffed at, that ended up flushing millions down the drain. That is all.

Permalink to Comment

8. nitrosonium on April 3, 2014 3:20 PM writes...

process patent = you can probably trust what is reported . otherwise (it has been our experience) much of the remaining patent literature is incorrect or misleading when it comes to reproducing results. most of those patents are not products from academic efforts.

Permalink to Comment

9. Am I Lloyd peptide on April 3, 2014 3:56 PM writes...

#6: I have met with Fukuyama a few times and agree that he is both incredibly smart and accomplished as well as a very nice guy. I do hope it's the co-author who botched up (well, ideally I hope nobody botched up but that doesn't seem to be the case here).

Permalink to Comment

10. Hap on April 3, 2014 4:02 PM writes...

@8: I thought the problem with patents was mostly that they didn't tell you everything you needed to know to repeat the work, not that they didn't tell you things that were wrong, but I haven't tried to do anything from a patent (reading them is bad enough). I assumed that a patent could be invalidated if the inventors were dishonest in it, but since that would take lots of money and time to prove, maybe they wouldn't care.

Permalink to Comment

11. MTK on April 3, 2014 4:44 PM writes...

The patent literature?

The game there is to tell enough to get a patent, but not so much that someone can actually repeat it.

That's completely different from the peer-reviewed literature which is to make sure others can repeat it.

Permalink to Comment

12. Anon on April 3, 2014 4:52 PM writes...

The digital age has made editing raw data too easy. When you had to photograph or photocopy your data it was much harder to fake these things. Combine this with the pressure cooker that many top labs are and we have the expected result. I'm impressed that Fukuyama is correcting this. Most labs avoid corrections like the plague.

When I was in grad school I knew some people that would do a final evaporation from carbon tet so that yields would get a boost without the telltale solvent peaks in the spectra. The smaller the scale you worked on, the more it would add fuzz to the results.

OTOH, I once reproduced >10 step synth of an unusual amino acid by Rappoport and got yields very similar to the reported.

Permalink to Comment

13. ....ymous on April 3, 2014 5:21 PM writes...

Don't you just love BMCL. A simple statement to the effect that the proton NMR and MS are consistent with the assigned structure is all you need.

Permalink to Comment

14. Anton on April 3, 2014 5:28 PM writes...

@13

If the reviewers demand to see the data, you can draw it with crayon on construction paper.

Permalink to Comment

15. NaturalChemist on April 3, 2014 6:44 PM writes...

@14
Good to see that our favorite misogynist is back at it again.

Permalink to Comment

16. jd on April 3, 2014 7:34 PM writes...

@4

thank you!

Permalink to Comment

17. Anonymous on April 3, 2014 9:40 PM writes...

@8 "patent literature is incorrect or misleading." The experimentals in a patent should always accurately reflect what the chemist did to make the compound, and therefore should be reproducible. That doesn't mean they have to be good chemistry. There is a requirement to disclose "the best mode", but, for example, if you're in med chem and you're only making a compound one time (and the vast majority of these compounds are only made once), however you did it becomes the best mode you have. That does not imply that it's anywhere close to the actual theoretical best way to make that compound. The goal for med chem is make as many analogs as fast as possible. Most of the routes haven't undergone any optimization and they can be really sloppy and roundabout ways to make the compounds. That's the nature of the work, and that's what the patents reflect. That doesn't mean they are not trustworthy, it just means that you can't expect them to be something they aren't. You have to read them as "some people made these compounds and this is how they did it" and then apply your own intuition and skill, like with anything, when deciding if and how to use this information.

Permalink to Comment

18. nitrosonium on April 4, 2014 9:02 AM writes...

every once in a while you will run across experimentals in the patent literature that are just wrong. wrong in either the fact that the experiment can't be re-done as written and/or the supporting data is wrong. that is all i am saying. none of the literature is perfect. it is just that journal claims are more frequently checked/reviewed than with patents. those are not usually scrutinized unless there is an attempt to invalidate a claim or the entire patent. all the academic recalls for questionable data is not an issue that is solely the domain of academic labs. when you read an experimental of any kind (except for Org. Syn.) it is a case of caveat emptor so-to-speak

Permalink to Comment

19. Cato the Elder on April 4, 2014 9:24 AM writes...

I applaud Fukuyama for doing these corrections, many profs might have been tempted to sweep this under the rug...

Permalink to Comment

20. Hap on April 4, 2014 9:31 AM writes...

No, anyone can be dishonest, I just figure that there's a particular incentive for total syntheses to be fudged because they probably won't be reproduced.

Permalink to Comment

21. See Arr Oh on April 4, 2014 10:51 AM writes...

@Cato The Elder: Exactly. Now, how to fix going forward?

Permalink to Comment

22. Cato the Elder on April 4, 2014 11:33 AM writes...

I'm not sure. I think this public "humiliation" (to be clear this is hardly that, I use this word for lack of a better one) will discourage others from following in the perpetrator's footsteps. Reviewers should always check the NMR supplementary information, but I'm aware many colleagues do not or only cursorily do so. However if these NMR alterations are done well, I'm not sure anyone could reliably detect them. Probably the only way would be to require the actual FID files to be submitted with publication. This does not seem such an onerous requirement.

Permalink to Comment

23. Anonymous on April 4, 2014 11:54 AM writes...

I am sympathetic to doing corrections rather than just retracting in this situation, because it appears there was one coauthor on nearly all of these that manipulated the data, and it doesn't appear to have affected the reproducibility.

Since science is not done in a vacuum, it is really, really awful for the other coauthors when one's hard work is retracted, even though the chemistry works, due to the stupidity of one bad apple.

Permalink to Comment

24. KSH on April 5, 2014 4:15 AM writes...

@23
I'm not convinced that the blame lies solely on one coauthor. Bearing in mind the recent corrections in Org Lett, manipulation of spectra appear to be a general problem in the Fukuyama group.

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
Employment Among New Chemistry PhDs
My Imaginary Friends Would Be Glad to Serve as Referees
Biopharma Stock Events for the Rest of the Year
A Drug Candidate from NCATS
No Scripps/USC
Studies Show? Not So Fast.
Outsourced Assays, Now a Cause For Wonder?
An Alzheimer's Blood Test? Not So Fast.