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

« Avandia: Going Under for the Third Time? | Main | More Belts Tighten »

December 7, 2007

Kids These Days!

Email This Entry

Posted by Derek

As pointed out in the comments to this post, the Menger / Christl pyridinium incident made the mainstream German media in the form of an article in Der Spiegel, which is about as mainstream as it gets. Even a news magazine with a page count as high as that one doesn’t cover organic synthesis very often, so I was curious to see how they handled it. Here’s my own translation of the last part of the article, after Christl sends along the bad news that Menger had rediscovered the century-old Zincke reaction:

Menger reacted reasonably, says Christl. “Fifteen minutes after my e-mail he wrote to me that he would immediately review things”. The American at first believed himself to be in the right: that he really had prepared a new 12-membered ring. “I hate to disappoint you”, wrote Christl, equally sure of himself – and referred to a mass-spectroscopic analysis of the products. These had given a higher molecular mass than a 6-membered ring would. But Menger had overlooked an important detail.

“I am no specialist in analytical methods, but I knew the mass spec method that Menger had employed clusters fragments together", says Christl. During the mass spec analysis, a 6-membered ring could appear to be a 12-membered ring. Menger did not know this effect – “A point we did not realize at the time”, he said remorsefully in the journal Nature. Christl reproaches Menger for imprecise work: “He should have asked a specialist in mass spectroscopy”.

Menger must now send in a correction to his work, and Zincke will be acknowledged. That puts Yamaguchi and Menger hard on Zincke’s trail, says Christl with amusement. Both of them used a particular salt, actually called Zincke salt, in their experiments to prepare the supposed 12-membered rings, but they were obviously completely clueless as to the origin of the name.

But another question remains: How could a 102-year-old reaction simply be overlooked, even though every article in a journal is proofread by an external reviewer?

Zincke had naturally published his work at the time in German. At that time, Germany was a center of research, and English was not yet the official language of science.

But was it really just the language barrier that made the entry into Zincke’s work difficult for Yamaguchi and Menger? Christl says: “This literature is so important, that it’s also given in English”. Maybe not Zincke’s complete original article from the “Annalen”, but at least a description of the reaction in the chemical handbooks. These remain in the libraries.

And here lies the problem, that hardly anyone just looks in the books, complains Cristl. “Young researchers just don’t get up any more from their computers. Most of them don’t even know that such a handbook exists”.

Lack of time and overloading of the reviewers are likely to blame. “They have to proofread a publication, bit by bit, every day. How is that supposed to work?” It seems, over and over, that old reactions are unknowingly rediscovered – because scientists simply don’t do their homework, says Christl. “For well-known scientists, the reading of the literature has become a luxury that they can no longer afford”.

I think we missed a golden opportunity in the penultimate paragraph to learn how to say “young whippersnapper” in German. Applying the term to Fred Menger does require a bit of an imaginative leap, admittedly. I’m glad that Spiegel turned down the chance to make this an “if only people knew German” article, though. But even if Christl’s points about literature searching are valid, I’m not sure that this case illustrates them.

I think this reaction would (should) have been picked up by newfangled tools available to those young ‘uns sitting at their desks. Can you get any more newfangled than Wikipedia? You don't even need SciFinder: a Google search for ("primary amine" pyridinium reaction) will give you this, which should be enough to follow up on. But as for hard-core literature searching, a few minutes of reading the pyridinium + amine literature should have turned up this from 1976 or this from 1970. But why go back that far? How about this review in Angewandte Chemie itself from those far-off days of. . .2006?

No, I don't think the problem here is that people don't know how to turn off the computer and go seek out the good ol' dusty handbook. Those have a lot of good information in them, certainly, and no one's ever cranked out better ones than the Germans have. But the problem here is that people didn't apparently didn't spend any time at all checking the literature. And what's more, the folks who send papers to Angewandte Chemie, and especially the ones who review them, don't even seem to be able to find a key reference published last year in the same damn journal.

Comments (20) + TrackBacks (0) | Category: The Scientific Literature


COMMENTS

1. MTK on December 7, 2007 9:37 AM writes...

To be honest, I don't see what the big deal is here. In my mind the system worked perfectly fine.

A) Someone finds something (or thinks they find something)
B) Sends it in for peer review (since this is a pharma / chem blog, let's equate that to a Phase I trial)
C) It gets published so that the wider scientific community sees it. (Sort of a Phase III trial)
D) Someone points out the errors
E) It gets retracted. (Never makes it to market)

Now we can all go back and Monday morning QB, but really, except for the faces that are wiping egg off, what really happened?

The real issue may be that most of us treat the literature like the Bible and expect it to be literal and infallible.

Permalink to Comment

2. Anonymous on December 7, 2007 10:10 AM writes...

Obviously ACIEE has published dubious quality research multiple times now. I think this should be time for careful introspection by ACIEE editors on how to salvage the reputation of their journal.

Permalink to Comment

3. LuckyChucky on December 7, 2007 10:11 AM writes...

Please tell me if this is the same F. Menger from Emory who spent much time in the early 1990's punching holes in the work of Breslow and Rebek...if so, oh the irony of it all!

Permalink to Comment

4. eugene on December 7, 2007 10:37 AM writes...

"And what's more, the folks who send papers to Angewandte Chemie, and especially the ones who review them, don't even seem to be able to find a key reference published last year in the same damn journal."

To be fair, Andjewandte doesn't have the pretty graphical abstracts that load up on the main page when you go to their website. You have to go to a special link which will show the graphics as very small pictures so you have to squint. What a pain! Or you can open up a PDF graphical abstract file, but you can't click on the pictures to get your article. Arghhh! The horrible inconvenience of it all!!! Is that journal run by old people who never watch TV and play video games!? I bet it is...

That's why I miss a week or two sometimes. JCAS though, has pretty pictures right away and if you click beside the picture, you get your article. It makes it easy and fun to check every week.

Permalink to Comment

5. weirdo on December 7, 2007 10:59 AM writes...

MTK,

You're missing the step between "2" and "3": somebody atually reviews the data from the Phase 1 trial. In this case, two reviewers absolutely blew it, and should be ashamed.

(I know, I know, they do it for free, they're busy, YADA YADA. But I bet they all put "Reviewer for Angewandte Chemie" proudly on their resumes.)

LuckyChucky,

Same guy.

Permalink to Comment

6. milkshake on December 7, 2007 11:01 AM writes...

Yeah, the system worked well - to the great embarassment of the authors, and Angewandte. If this happened with the Annalen of Albanian Chemical Society it would be less of a story.

I really think that the ease of use Beilstein and Scifinder makes people more lazy and superficial in their lit searches. I remember in 80s it took about 2 days in library to check on research project idea. I would go through the Beilstein books and Houben Weyl, OrgSyn and Org Reactions in our uni library, then take off to the sole institute in Prague that had the 5-year condenzed Chem Abstract indexes. (using the semi-anual CA indexes to cover last 70 years is only for a masochist perv)

I am not eager to return to bad old days but when one is forced to plod through countless pages of keywords in General Subject or entries in General substance indexes, one often finds all kinds of useful stuff that one is not looking for.

Permalink to Comment

7. Hap on December 7, 2007 11:07 AM writes...

MTK - I think that it does point out a systemic flaw, though not in the ability of the system to self-correct. People expect the things they read in journals to be, if not perfectly true, then as good as they can make them at the time with what they know. (Particularly with how much people/institutions pay for them - it isn't unreasonable to expect journals with $10-20K subscription fees to do what they are paid to do.) If people didn't bother to check the literature for previous examples while doing research or reviewing it, and the results are publishable in one of the best journals in chemistry, how much of lesser stuff is not as accurate as we think it is? Other publications (newspapers) aren't expected to be absolutely correct - but they are expected to do their research and check their sources insomuch as they can.

There is also the problem that if we can't necessarily trust what research is published, we can't necessarily trust that good things haven't been rejected from publication for similarly bogus reasons. That might be the bigger problem - we can correct what we see that is wrong, but we don't know what we don't see that is correct. (It could have been published elsewhere, but the same reliance and author and reviewer skill and work exists for other journals.)

Permalink to Comment

8. eugene on December 7, 2007 11:08 AM writes...

"I am not eager to return to bad old days but when one is forced to plod through countless pages of keywords in General Subject or entries in General substance indexes, one often finds all kinds of useful stuff that one is not looking for."

Have you tried the new SciFinder? I bet you have. The new 'similarity' search is awesome and is just what you were looking for. When you look at compounds that are 80% percent similar, you find a lot of stuff you weren't expecting but that is very interesting. I love it personally. I know it has helped me out with a synthesis or two and made me think of a project in an entirely new way after looking at those 60-80% similar structures.

Now, they just need to implement this feature for keywords...

Permalink to Comment

9. MTK on December 7, 2007 12:36 PM writes...

weirdo,

I include those folks (the in between step you mention) amongst ones wiping egg.

Hap, what you mention is the real issue. Have our expectations out of journals, referees, and the literature become unrealistic? I remember going to a symposium once which included biologists and chemists, and one of the biologists said to me afterwards, "You guys must really trust each other." His comment was directed at the fact we rarely present chromatograms, spectra, or any other raw data. At least the biologists show their gels.

You then point out that if we can't trust what's been reviewed and published then we can't trust what's been reviewed and not published. Let me take another angle at that: Maybe the fact that we have such high expectations of the literature prevents stuff from being published if editors and reviewers become too cautious. So if the community as a whole would set lower expectations than, yes, more "incorrect" things would be published, but perhaps more inventive things also. In either event these could be verified / debunked by the community as a whole.

Permalink to Comment

10. Hap on December 7, 2007 1:04 PM writes...

I think that there are a variety of journals for research to allow outlets for research that might be speculative, imperfectly reproducible, have a narrow substrate range, etc. If reviewers are unable to make these distinctions effectively, then the distinctions of journals become useless - you don't know what is good and bad in a given place, and no one has the time to cover it all. You don't brand speculative research as secure, or a reaction useful in a few cases as a reaction with wide and unbounded utility. There probably aren't enough places to report failed reactions or really speculative research, though.

If the distinctions beween journals and levels of reliability/quality in research becomes useless, then you have a problem - there is too much research to read oneself, and most of it won't be read, let alone tried and verified/rejected, at all. If someone doesn't get the reported result, they might figure that it's their own fault and not that of the original authors. Only rarely are papers disproving a method or retractions given, so what you would be left with is a morass of articles, among which people can't make useful distinctions or assumptions of truth. It would be preferable for the journals to provide reliable estimates of value (that is, after all, their job) than to throw everything in piles and hope that it is all read and processed appropriately.

Permalink to Comment

11. Roadnottaken on December 7, 2007 2:55 PM writes...

Don't forget that 99% of what's in the literature is probably pretty solid. That an error gets through the review process from time to time is not an indication of a system-wide failure, IMHO. Lesson learned: think critically and carefully about everything you read and don't believe surprising results until they've been reproduced a few times.

Permalink to Comment

12. milkshake on December 7, 2007 9:59 PM writes...

..not realy. The publications published by all-Indian groups in Indian journals are quite terrible. There is also lots of crap in patent literature. More often than not I find that completely half-assed procedures with miscalculated or missing ratios of reactanta, with irreproducible or missing workup /purification procedure make it into patent examples. You are lucky if it is not an outright "prophetic example".

Permalink to Comment

13. Petros on December 8, 2007 8:56 AM writes...

Like Milkshake I too remember trawling through CA 5 year indexes and occassionally old volumes of Beilstein. The largest problem with the latter, although my German isn't great, was the arcane indexing and classification style which was hard to get to grips with.

Apropos Milkshake's comments on experimental details and procedures. It's not just low grade journals. Just try replicating the reported yields of a major, US-based, Nobel laureate's group! And they frequently resort to not reporting the detailed protocols (or used to)

Permalink to Comment

14. Jose on December 8, 2007 12:19 PM writes...

My personal favorite is a methodology group which consistently publishes 100% yields, off a column, in duplicate.

Permalink to Comment

15. Wavefunction on December 8, 2007 6:53 PM writes...

Menger's jabs at Breslow, Houk and others in the 80s and 90s had some pretty valid points in them. It's worth taking a look at the original papers.

Permalink to Comment

16. PG on December 10, 2007 6:54 AM writes...

So the system has worked, huh?
Angewandte is supposed to have very high standards, which include originality. Therefore there is an extra burden on the referees, who have to not only make sure the science is sound, but also that it is original.
In this case, either the referees decided that they had enough expertise in the field and did not need to carry out simple (as Derek proved) literature search, or they did not bother at all.

Are 'good' referees being sent too many manuscripts for the system to work properly? Then the journals need to spend more money on professional editorial staff to be more selective, AND authors need to stop sending average papers to the top journals.
Oh, and the publishers (and the scientists who agree to be on their editorial boards) should think twice before creating new journals.

Permalink to Comment

17. MTK on December 10, 2007 12:51 PM writes...

Yes, PG, the system worked.

Maybe further down the chain then you would like, but it worked.

Let me add this, do you think that after a second highly publicized instance that Angewandte and it's editors will be more circumspect going forward? Do you think that maybe the referees are embarrassed and will do a better job in the future? Do you think that all of us will do a better job in reporting and refereeing?

I think we can all reasonably answer "yes" to those.

Permalink to Comment

18. Hap on December 10, 2007 1:20 PM writes...

I think that I would answer "yes", particularly with the letter, but I would have figured that after LaClair's hexacyclinol/deoxoudol paper (with no synthetic procedures or spectra of intermediates) ACIEE would have strengthened their data requirements. No dice there.

I don't what leverage jourmals have to improve refereeing, particularly with the tightrope that I've heard about for referees (if they are too severe with papers rather than biased or mean, they don't get to review any more, and if they are lax, as here, bad things happen). I don't know how much work being an editor is, but expecting editors to do more of the referees' job seems like a difficult sell.

In this case, I don't think correction after the fact helps (though it is much better than nothing) - if I wanted a grab bag of uncertain chemistry, there are lots of cheaper places to get it (*cough* patents *cough*).

Permalink to Comment

19. Dave Eaton on December 14, 2007 1:18 PM writes...

how much of lesser stuff is not as accurate as we think it is?

There's probably a sliding scale, depending on the number of eyes on the work.

I agree that perfection is the goal, and short of that, proper care and self-scrutiny by the authors, backed up by due (or even some) diligence is what we should demand.

But this is the system working. It would be grand if people never got lazy because we are all such dedicated, perfectly rational scientists. But we are not. So the fear of derision and ridicule by peers is the quality assurance regime that we actually use. If it never was invoked, the system would be gamed a lot more often, I'd bet. Goofing up gets you laughed at. Lying gets you kicked out. Fear is a great motivator to get people to act right.

Permalink to Comment

20. probatefox on January 9, 2012 2:28 PM writes...

Poor Martha, Im afeared, was never meant to be successful at running for anything ceptin maybe a bus. Fact is, shes made such a mess of things that yesterday her doctor diagnosed her with a rare form of Bidens foot in mouth disease. Maybe shell get some votes from people who feel sorry for her although I feel more sorry for the reporter that her campaign manager beat up last week and kicked down some subsay stairs. Anyway, thats how I heared it. But thats just me.

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
Gitcher SF5 Groups Right Here
Changing A Broken Science System
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