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

« What's the Condensation Record? | Main | India's Research Culture »

May 25, 2010

A Word to the Wise

Email This Entry

Posted by Derek

Today's entry on an embarrassingly wrong structure in Bioorganic and Medicinal Chemistry Letters has a number of people in the comments talking about experiences they've had reviewing for the journal. I've done reviews for them myself, naturally, as well as for other journals. I have too few examples to judge from, but (so far) I have managed to kill off papers at other journals, but I have never managed to kill off a manuscript at BOMCL. I have - apparently like some other people - recommended in the past that a paper be published (if at all) only with major revisions, only to see it sail through basically untouched.

I hope I'm not being unfair here, because there are a lot of hard-working people at the journal. And similar stories can, I'm sure, be told about every other journal (in every other discipline). But I think that BOMCL gets so many manuscripts that their workload is very high. Unfortunately, the journal also publishes a great deal of what it gets. We're unlikely to see the real figures any time soon, but I'd have to guess that the percentage of papers rejected is definitely lower than average.

I also realize that I'm open to accusations of conflict of interest here, since I'm on the editorial board of a competing journal, ACS Medicinal Chemistry Letters. What I can promise is that I will, in fact, work to keep any papers that I consider inadequate out of those pages (and, at the same time, to encourage good work to go there). I've tried to do that with BOMCL, too, in my capacity as reviewer, but it just hasn't always worked out.

But perhaps this is something I can do for them: to point out, publicly, that their credibility as a venue for medicinal chemistry results has suffered recently. When people get the impression that work is being sloppily reviewed at a given journal, they wonder how much they should trust the other papers that get published. Bioorganic and Medicinal Chemistry Letters has been around for twenty years now, and has published some very useful stuff over the years. There's a real place for it in the publishing world. But it's been better than it has been recently, and it should be better than it is. I hate to say this. But someone should.

Comments (35) + TrackBacks (0) | Category: The Scientific Literature


COMMENTS

1. john on May 25, 2010 1:04 PM writes...

This is a major issue, peer review is important and too many people just ignore the responsibility.
If the peer reviewers the Lancet had forwarded the paper to maybe 25% of american parents wouldn't think that vaccination causes autism.

Permalink to Comment

2. Evorich on May 25, 2010 1:06 PM writes...

Very polite Derek - well done.

You suggest that they get so many papers and their workload is so high, yet the fact that they publish things that referees recommend that they should not be would suggest the opposite of this, would it not.

It also seems odd that if they go through the work creating process of getting papers refereed, why not use the results of that process?

Permalink to Comment

3. J-bone on May 25, 2010 1:10 PM writes...

Tetrahedron Letters has also taken a similar dive in quality. Surprising to compare its content now to Corey's papers from the 70's.

Permalink to Comment

4. partail agonist on May 25, 2010 1:32 PM writes...

I've reviewed a dozen or so, and most of the ones I flat-out rejected did not get in. The ones I asked for major revisions seem to get in without as much change as was needed.

Most of the ones I flat-out rejected happened to be pretty unreadable, though, even considering the authors were non-native English speakers.

A scientific paper should not sound like dialog from the Simpsons (I'm thinking especially of the guy -Apu?- at the Quick-E-Mart who sells Squishies). I'm sorry if that sounds insensitive, but PLEASE have a native English speaker look over your manuscript. It is hard enough to go through the science without explaining the basics of sentence structure in my review.

/rant

Permalink to Comment

5. I can spell partial on May 25, 2010 2:51 PM writes...

My experience with BOMCL from both sides of the reviewing process has been quite positive. Of four papers I submitted, each required only minimal revision and all critical comments seemed reasonable to me. For the one paper that I reviewed for BOMCL, it needed a moderate amount of revision. All my recommended revisions and corrections were done before it was published.

Permalink to Comment

6. Wagonwheel on May 25, 2010 2:57 PM writes...

This one has still not been corrected...

http://pubs.acs.org/doi/full/10.1021/cn100022a

Permalink to Comment

7. Petros on May 25, 2010 3:01 PM writes...

I guess I've been lukcy, tos I've reviewed have needed, at most,minot tweaks.

But with two new rapid Med Chem Journals,from the ACS and (shortly) the RSC, they do need to sort it out.

It is my impression that Tet Letts is much poorer than it used to be but, as some comments have alluded to, even JACS has had some issues with quality in the past

Permalink to Comment

8. bbooooooya on May 25, 2010 3:06 PM writes...

Does anyone really take BOMCL (or Tet. Lett. post ~1995 or so) seriously?

I recall from the turn of the century at TSRI that for, let's say, "second rate" work that all one had to do to get a paper into BOMCL was to "walk it down to Dale's office".

Permalink to Comment

9. Sili on May 25, 2010 3:22 PM writes...

If the peer reviewers the Lancet had forwarded the paper to maybe 25% of american parents wouldn't think that vaccination causes autism.
I'll have to admit that I haven't read the original paper, but my impression is that it was not supported by evidence (and Wakefield has now been struck off because he failed to obtain ethics clearance - not that he would have received that even if he'd applied, given the utter lack of evidence). The 'problem' with peer review is that assumes honesty, so peer reviewers are godawful at catching outright fraud.

And then of course there's The Journal of American Physicians and Surgeons where fraud seems to be a requirement for publication.

Permalink to Comment

10. john on May 25, 2010 4:30 PM writes...

Sili
The original paper has some serious issues which probably shout have been apparent, of course there is a significant media fault (apparently if you pose in playboy you are more able to read and understand the literature than the majority of the scientific community).
What does everyone think of the Medical Hypothesis Journal issue that have come up, where Elsevier is forcing the journal to go to a peer reviewed system.

Permalink to Comment

11. Chemjobber on May 25, 2010 5:47 PM writes...

Layoffs at Albany Molecular:

"As part of cost reduction efforts, AMRI is reducing its US workforce by about 10%, or an estimated 80 jobs. This includes currently open positions that the company is not filling at this time."

Article linked in my handle.

Permalink to Comment

12. Acid hydrolysis on May 25, 2010 7:12 PM writes...

Don't write off BMCL so easily. For instance many of the initial syntheses papers for best-selling drugs like kinase inhibitors were published in BMCL

Permalink to Comment

13. BioBrit on May 25, 2010 7:26 PM writes...

I'm going to buck the trend here and say that I like BMCL, and have published a few papers there. JMC is great, but my feeling has always been that work to get published there has to be high quality AND groundbreaking. So, you can publish the very first series for a target there, but the second and third generation series won't get it, even if the work is of excellent quality. That work deserves to be published somewhere. I guess my perfect world would have BMCL publish everything that was good quality, not just everything.

As to the thiophene paper in question - I only glanced at the abstract. Is the data bad as well, or just the structure (and did the actually maintain that was the structure, or was it a bad drawing error carried through? It's certainly not excusable to miss that obvious of a drawing error, but a lot more so than actually proposing that as a structure.

Permalink to Comment

14. Hahaha on May 25, 2010 7:46 PM writes...

I guess I've been lukcy, tos I've reviewed have needed, at most,minot tweaks.

Did this guy just prove the point or what?

Permalink to Comment

15. BOMCL author on May 25, 2010 8:13 PM writes...

For what it's worth I published in BOMCL six months ago and felt the review system worked pretty well.

Our manuscript was reviewed by two separate reviewers who did a good job of pointing out 2 minor errors in reference compound structures and making a good number of valid comments and worthwhile suggestions.

The manuscript was accepted with minor revisions (which we included in their entirety) and published rapidly.

Permalink to Comment

16. BOMCL author on May 25, 2010 8:17 PM writes...

I should also note that I am working on another BOMCL manuscript at this very minute and that I am suddenly feeling less sanguine about my target journal choice after reading Derek's post and some of the comments...

Permalink to Comment

17. Rock on May 25, 2010 8:57 PM writes...

BOMCl to me has perhaps the widest range of quality than any other journal I can think of. Yes, there are some very good papers in there, however, there are also too many really bad papers. I will repeat what I posted in yesterday's blog: Scrolling through the abstracts on your RSS reader can be a useful learning tool for younger medicinal chemists to be able to be able to differentiate the quality from the crap. I will say that quite often, the crap comes from academic labs.
I wonder if there is a correlation between the drop in quality in journals and the timing of the editorial change that required authors to submit the names of their friends..er I mean potential reviewers....

Permalink to Comment

18. Anonymous BMS Researcher on May 25, 2010 9:22 PM writes...

Back in graduate school my major professor once got the same paper for review from journal after journal; after about five of them he started sending very brief reviews, a slightly more polite version of "Dear Editor, I have recommended seven journals reject this piece of crud, and wish the authors would give up instead of journal shopping."

Permalink to Comment

19. Black tar again on May 25, 2010 11:14 PM writes...

This is, sadly, yet another price we are paying for metric-driven assessment of publications, predominantly for academics. Tetrahedron Let used to be a good journal until the lazy, the fraudulent, and the desperate twigged to the idea that you can publish there without much experimental data to speak of. Bit by bit the second-rate dross started to be dumped into TL until it became close to worthless. BMCL is heading the same way.

At the other end, the top-shelf journals are becoming too much of a hassle to publish in (especially for us industry people with less spare time on our hands *ducks*). A recent JACS paper of mine was a 20 page manuscript with 150 pages of supplementary info!

Sometimes I despair for the future of scientific publication. It used to be mostly about communication of results but now it's all a game of citation rates, h-factors and impact factors. Witness the proliferation of reviews in previously article-only journals, blatant self-citation and "citation clubs", journal shopping to try to get into the highest impact journal rather than most appropriate journal... all very sad.

Permalink to Comment

20. Petros on May 26, 2010 2:13 AM writes...

For medicinal chemists with work that their companies allow them to publish there has always been a very limited choice of journals

J Med Chem was preferable but its insistence, until recently on CHN analyses, was a significant barrier
Chem Pharm Bull is rarely used by non-Japanese groups
Arzneimittel Forschung was only used by some German-based groups
Eur J Med Chem is used by ?????

That left BMCL as the preferred route

Permalink to Comment

21. Ben Zene on May 26, 2010 7:57 AM writes...

I spend a lot of my time reading pharmacology papers. If you think that journals such as BOMCL are bad then you should try looking at chemical structures as drawn in these journals. Stereochemistry is usually non existent and heteroatoms such as N, O and S often omitted. It drives me crazy trying to make sense of it....

Permalink to Comment

22. bbooooooya on May 26, 2010 8:02 AM writes...

"J Med Chem was preferable but its insistence, until recently on CHN analyses, was a significant barrier"

A significant and important barrier to keep out garbage such as that published in TL or BOMCL. Lowering of standards to reflect laziness is really not the route to better science, though perhaps it partly explains the poor performance of big pharma.

I hated having to get EA in grad school (my adviser insisted on it, even for stuff in sub-par journals), and would have been much happier to rely just on HPLC and/or NMR---both are much more forgiving in terms of hiding impurities.

Permalink to Comment

23. BioBrit on May 26, 2010 10:21 AM writes...

""J Med Chem was preferable but its insistence, until recently on CHN analyses, was a significant barrier"

A significant and important barrier to keep out garbage such as that published in TL or BOMCL. Lowering of standards to reflect laziness is really not the route to better science, though perhaps it partly explains the poor performance of big pharma."

Are you claiming that if big pharma routinely did EA then they would have done a better job? I think not. The lack of EA is less one of laziness, and more one of lack of relevance - if it isn't that active/stable/bioavailable, who cares?

Having said that, effort/cost does play a role, its a lot easier to walk your sample down the hall at TSRI than to send it off to an external lab, deal with your company's legal group to get authorization to ship, ......

As a community, we can't have it both ways. We want everything published so that it is out there, yet we expect our journals to maintain high impact factors and not publish the non-results and the boring. It is akin to the clinical trials world, people want all clinical trial data published, regardless of the outcome, but the journals are pushed to not include "Study Fails To Demonstrate A Damm Thing" papers. I'm no MBA, but I'd guess there isn't a good business model behind "The Journal Of I Made Stuff That Was A good Idea But Didn't Quite Pan Out".

Which of course doesn't mean the journal should publish BAD science, just BORING science.

Permalink to Comment

24. Mike W on May 26, 2010 10:27 AM writes...

I have also reviewed manuscripts for BMCL that I spiked but that later got in amyway. I find this unacceptable. Hwoeever, if the other reviewer accepts it with revisions the manuscript will often be sent to a third - tiebreaker- referee, which can lead to publication. If so, the problem is with the reviewers.

I have also seen a number of papers in J Med Chem that are, to put it bluntly, so minimal in import that they would be tough calls for me to put in BMCL. I am fairly sure that the problem in those cases is the reviewers. Scientists, when acting as reviewers, need to take their role as guardians of the journal's quality seriously. Editors can and should not allow many manuscripts to even reach the desks of reveiwers, but this is in many cases too much to ask of individuals who have a lot of manuscripts crossing their desk. Also, I have spoken with editors who have said they sometimes like to send bad papers through to reviewers for a more definitive smack down, hoping to send a message to the author and his co-workers, and thereby, hopefully, the entire scientific community. We could argue the wisdom of that, but regardless, that scenario just makes the reviewer's job even more important.

Certainly the editors of BMCL have a lot to answer for, but if reviewers are tough enough and insistent enough, editors may possibly be brought in line. Next time you spike a paper and it gets published anyway, send an angry email off to the editor who handled the ms.

Permalink to Comment

25. Mike W on May 26, 2010 10:34 AM writes...

And to 22, inclusion of EA data for bioactive molecules isn't rigorous science, it's pedantry. There are better ways to judge a compound's purity, and given the error bars associated with most assays, in vitro data simply doesn't require extreme purity. Compounds that go further will always - jeez, I hope this is true of your shop - be made in super-pure form.

Permalink to Comment

26. Will on May 26, 2010 10:36 AM writes...

@23

I had also pined for a journal like you describe, maybe an online repository where people could simply upload failed experiments, so as to keep others from re-inventing the same square wheel.

Now, I think it's a non-starter, why publish something that didn't work out (that your group/company has presumably abandoned) and let the competition take a swing at it? From a business sense, it makes far more sense to keep all the knowledge internal.

Permalink to Comment

27. BioBrit on May 26, 2010 10:56 AM writes...

@Will (26)

I suppose it depends where you draw the line at "didn't work out". Weren't active? Were active in-vitro but not in-vivo? Were active in-vivo but program got killed anyway? Failed in Phase 1?2?3? If we only published stuff that made it to market, the journals would be very quick to read.

I think we, as scientists, all still want to publish as we, as hiring managers, all expect to see a solid publication record. And at some level, even if it doesn't make perfect sense, business support it, perhaps because enough of the management are, or were, scientists too.

I've always thought industrial medchem papers/posters fall into 2 categories. 1) Stuff that is far enough through the clinic to be safe to talk about, and 2) Stuff that sounded like a good idea but got killed, so it doesn't matter to talk about.

Permalink to Comment

28. Hap on May 26, 2010 10:56 AM writes...

I don't agree that EA is pedantry.

First, if a compound goes further than usual for JMC, it won't be in JMC, but probably won't be published (at least for a long time), and if the compound bio data isn't what was first said, it won't generally be mentioned again (rather than being corrected), so that something wrong could linger forever. Hence, if you ascribed bio data to the wrong compound, because it wasn't pure enough, no one will ever know (unless it's in a bustling field, and someone else does it).

Second, I figure that people would like to know that the bio data claimed isn't simply an artifact of their synthesis (a byproduct isn't giving the observed data) - MS only tells you that you've made the correct compound, not that it's actually pure, and NMR could be missing up to 3-5% of a byproduct, plenty to give a result.

If I'm actually saying that a compound has the data I claim in public (with my name, or close enough, on it, and in a forum where I am actually expected to be accurate), I sort of figure that I'd want to know that sort of thing, anyway.

Permalink to Comment

29. Hap on May 26, 2010 11:11 AM writes...

As a sort of correction, I don't think JMC requires EA anymore - I think you can use HRMS and purity specs (TLC in multiple systems?) instead of EA for compounds there.

Permalink to Comment

30. Aspirin on May 26, 2010 12:25 PM writes...

These are anecdotes, not evidence. Those here who are chagrined that BMCL published papers in spite of their rejections and reviews, consider the simple fact; perhaps the other two reviewers actually gave the manuscript good reviews? In such a case the editor would simply listen to the voice of the majority, nothing wrong in that.

Permalink to Comment

31. bbooooooya on May 26, 2010 1:19 PM writes...

"Are you claiming that if big pharma routinely did EA then they would have done a better job? I think not. The lack of EA is less one of laziness, and more one of lack of relevance - if it isn't that active/stable/bioavailable, who cares?"

No, I am claiming that if big pharma employees worked more diligently (and got proper characterization) that the industry would be more successful. Not saying one ought to get ea for useless cpds, but for something moving into animals, certainly.

Don't waste pure thoughts on impure compounds. if there is a more rigorous way to establish purity (i.e., fish out impurities that don't have easy to see chromophores or nuclear active nuclei) please let me know (maybe a weight based HPLC assay). Note, this clearly would not apply to biologics.

"Having said that, effort/cost does play a role, its a lot easier to walk your sample down the hall at TSRI than to send it off to an external lab, deal with your company's legal group to get authorization to ship"

Bollocks. Saving a bit of time to package sample, and a day or 2 to fedex is cheaper that injecting impure cpds into animals.

Permalink to Comment

32. Will on May 26, 2010 4:05 PM writes...

BioBrit

My wish for a repository of failed experiments was from my grad student days doing tot. syn & glycosylation methodology. Could spend months trying to get a particular coupling to go, eventually fail and move on the something else. Then, 2 years down the road discuss it with another group, which then says they tried the same thing 5 years ago, also couldn't get it to go. Maybe if I'd seen their data to begin with, I wouldn't have burned months.

Then again, if nobody took cracks at something that had failed with someone else, science would not be so far along.

Permalink to Comment

33. BioBrit on May 26, 2010 4:27 PM writes...

@Will

I think I couldn't be bothered to write up failed experiments, as in I didn't manage to get them to go. I do see BMCL as the rightful repository for the second or third generation of a SAR campaign that didn't perhaps make any great strides forward in potency, etc but maybe was a new scaffold. Good for people to know, but didn't really get you further towards a drug.

JMC should hold the first series for an orphan target, major breakthrough in potency, stability, major new structural class, or we got to the clinic with this one.

Agree that too much information on what failed might inhibit investigation.

Permalink to Comment

34. Pat Pending on May 26, 2010 9:51 PM writes...

Bbooooooya, two comments 1) At my company you have to get Legal to sign off on any shipments and I will be quizzing you, and if I don't like the answer your boss, on whether this shipment is really necessary 2) Again at my company, all manuscripts must be reviewed by a patent attorney before being submitted for publication and if I see potential patentable subject matter such as the crystalline form of a compound because a patent directed to a crystalline polymorphic form of a drug or precursor I will not authorize disclosure until after the compound fails in the clinic and the project is canceled or patents directed toward the crystalline form are filed.

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
A Last Summer Day Off
The Early FDA
Drug Repurposing
The Smallest Drugs
Life Is Too Short For Some Journal Feeds
A New Look at Phenotypic Screening
Small Molecules - Really, Really Small
InterMune Bought