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

« Podcast on Avastin | Main | Lipitor Expiration Day »

November 30, 2011

Finding Even More New Reactions By Looking For them

Email This Entry

Posted by Derek

How do you find new reactions? I blogged here in September about a very direct way of doing it, from John Hartwig's lab: set up a bunch of things and see what happens. I liked it very much, but opinions in the comments were mixed. Some people found this approach refreshing, while others found it more simplistic than simple.

Well, get ready for some more, courtesy of the MacMillan group at Princeton. This paper has just come out in Science on reaction discovery, and it takes a very similar approach to "accelerated serendipity". They were looking at photoredox catalysts, which have been used for some interesting studies in the past few years. You mostly see iridium and ruthenium catalysts, with variations of tris-bipyridyl ligands on them, but the variety of reactions that they can initiate is extraordinary.

Clearly, there must be a lot of reactions in this area that haven't even been found, and that's what this latest paper sets out to do:

Assuming that serendipity is governed by probability (and thereafter manageable by statistics), performing a large number of random chemical reactions must increase the chances of realizing a serendipitous outcome. However, the volume of reactions required to achieve serendipity in a repetitive fashion is likely unsuitable for traditional laboratory protocols that use singular experiments. Indeed, several combinatorial strategies have previously been used to identify singular chemical reactions (2–11); however, the use of substrate-tagging methods or large collections of substrate mixtures does not emulate the representative constituents of a traditional chemical reaction. On this basis, we posited that an automated, high-throughput method of reaction setup and execution, along with a rapid gas chromatography–mass spectrometry (GC-MS) assay using National Institute of Standards and Technology (NIST) mass spectral library software, might allow about 1000 random transformations to be performed and analyzed on a daily basis (by one experimentalist). Although we recognized that it is presently impossible to calculate the minimum number of experiments that must be performed to achieve “chance discoveries” on a regular basis, we presumed that 1000 daily experiments would be a substantial starting point.

That it would, and by combining a broad selection of interesting starting materials with several plausible photoredox catalysts, and then basically just letting things rip, they found one. Dicyanobenzene, as it turns out, does a radical coupling with tertiary amines, giving you a direct C-C bond formation route that arylates next to the nitrogen. It's a perfectly believable reaction, but there are a lot of perfectly believable reactions that you could draw in this area that don't actually work.

Looking over the paper, it appears that the more time-consuming parts of the experimental setup were avoiding known chemistry in the starting combinations, and looking over the results to see what was worth following up on in more detail. Those are both human-brainpower intensive tasks; the rest was automated as far as possible. Interestingly, it appears that MacMillan had earlier been trying a very similar approach to that Hartwig paper I blogged about in September, doing reaction discovery with transition metals. But they then switched to photochemistry, thinking that this might be a more wide-open field.

It's not like the reaction dropped out of the robotics fully formed. They saw a new product form with an iridium catalyst, dicyanobenzene, and N,N-dimethylaniline, but further optimization gave better (and more general) conditions. That's as it should be; there's no way (yet) to run enough experiments to both find new reactions and the best ways to run them in one shot. But just getting a whiff of something new and useful is enough, and I don't see any reason not to engage in automated searches for such things.

But from the reaction in the comments here to that Hartwig paper, I gather that not everyone agrees. As far as I can tell, one objection is that famous talented organic chemistry professors shouldn't have to engage in such brute-force exercises. The more elegant way to come up with these things, by this opinion, is to use more brainpower up front, rather than just mixing up a bunch of stuff to see what works. I suppose - not being a famous talented organic chemistry professor, myself - that I'm not so proud. But then, John Hartwig and Dave MacMillan are FTOCPs, and they seem to have swallowed their pride enough to find something new. And good for them!

Comments (45) + TrackBacks (0) | Category: Chemical News


COMMENTS

1. Morten G on November 30, 2011 9:15 AM writes...

Pretty much like macromolecular crystallography. Serendipity with a sprinkle of canniness and a lot of optimization (sometimes very little though).

Permalink to Comment

2. Colm on November 30, 2011 9:43 AM writes...

What is the safety profile of experimentation like this? Granted I am not a chemist in any form, but it seems to me that the sentence:

Assuming that serendipity is governed by probability (and thereafter manageable by statistics), performing a large number of random chemical reactions must increase the chances of realizing a serendipitous outcome.

Could just as easily end 'or an explosion.'

Or are they confining these types of studies to exploring generalized reaction types unlikely to react violently?

Permalink to Comment

3. Nick K on November 30, 2011 9:56 AM writes...

Colm: The likelihood of an explosion is governed by the difference in enthalpy (delta H) between the products and the reactants. The greater the difference the more exothermic the reaction, and hence the more likely there will be an explosion. The reactions explored in this work probably have small or even negative delta H's, and are thus very unlikely to be explosive.

Permalink to Comment

4. Curious Wavefunction on November 30, 2011 10:02 AM writes...

Brute force and serendipity are fine. Supramolecular and solid-state chemists have been taking advantage of this approach for a long time - no painstaking bond-by-bond design, just mix in a bunch of chemicals and see what happens. It's time the synthetic chemists learnt from their solid-state brethren.

Permalink to Comment

5. will on November 30, 2011 10:03 AM writes...

I'm sure that one of the original FTOCPs Sir Robert Robinson would have scoffed at using spectroscopy and crystallization to determine an NP structure in less than a year, when careful degradation studies performed over a decade+ could yield similar information with much greater ~intellectual~ effort

Obviously that's tongue in cheek, but we should remember that the reason we (taxpayers) fund FTOCP is to actually discover stuff, and not simply engage in exercises that make themselves appear intelligent.

Plus, I would imagine this technique would serve a a leaping off point, or lead indentification. I'm sure there's plenty of opportunity to optimize the cyanobenzene catalyst using "conventional" brainpower

I imagine explosions are not much of an issue if the amounts of reagents are kept small (MS-GC analysis requires only a tiny amount of material) and proper shielding/safety precautions are employed

Permalink to Comment

6. Nick K on November 30, 2011 10:16 AM writes...

Correction to my previous post: exothermic reactions have NEGATIVE delta H's (i.e. they emit heat), so please disregard the the part which says "or even negative delta H's". Apologies.

Permalink to Comment

7. simpl on November 30, 2011 10:17 AM writes...

Reminds me of the 4-colour theorem. For a purist it is disappointing that a solution took 50 pages and computer grunt insead of elegance. The printer, though, is happy to know that he only need order four inks.
Now, who has a list of reactions waiting for a solution?

Permalink to Comment

8. Curious Wavefunction on November 30, 2011 10:31 AM writes...

4-color theorem: It's worth noting that two nineteenth century English mathematicians published "elegant" proofs of the theorem that eventually turned out to be wrong. In chemistry as in math, elegance is no guarantee of "truth" (which in chemistry would really correspond to "utility")

Permalink to Comment

9. luysii on November 30, 2011 10:34 AM writes...

Having plowed through 3/4 of Anslyn and Dougherty, and about to start the chapter on calculational organic chemistry (COC), this sort of experimental work gives me the creeps (about COC). Is it just being used to justify what we already know (post hoc propter hoc), or can it actually predict things? Shouldn't COC have figured out all these newly found reactions?

I have similar misgivings about protein structure prediction -- see http://luysii.wordpress.com/2009/11/29/time-for-the-glass-eye-test-to-be-inserted-into-casp/.

Hopefully all will be revealed before the end of the year. Anslyn and Dougherty have been great so far (except for one chapter)

Pleased to see that MacMillan and Hartwig both have ties to the PU chemistry department.

Permalink to Comment

10. Folger on November 30, 2011 10:40 AM writes...

One has to wonder how junk like this gets published in any journal much less Science.

Permalink to Comment

11. old man on November 30, 2011 10:56 AM writes...

This should be viewed as an admission of defeat. Just turn the brain off and let the equipment do the work. This is just like combichem in the pharma industry 10 years ago, and we know how well that worked out.

Permalink to Comment

12. J.S. Boc on November 30, 2011 11:09 AM writes...

I think part of being an FTOCP is recognizing problems that don't have good and general solutions and coming up with them, elegant or not. I don't see how "accelerated serendipity" can replace that. Nevertheless, I can see its use for discovering some interesting chemistry.

Permalink to Comment

13. Will on November 30, 2011 11:30 AM writes...

Y Kishi is a FTOCP, but the Ni-Cr coupling was a surrepticious discovery. Much of the apparent low hanging fruit in organic chemistry has been plucked, and nowadays you find groups dedicated to modifying ligands to trivially increase yields of archetypical reactions

If the combi-reaction scheme identifies new trees from which to pluck fruit, I say great. Who cares it didn't come from a dream or whiskey soaked inspiration?

Permalink to Comment

14. Paul on November 30, 2011 11:50 AM writes...

"Civilization advances by extending the number of important operations which we can perform without thinking about them." -- Alfred North Whitehead

Permalink to Comment

15. Mr. Fixit on November 30, 2011 12:03 PM writes...

If we step back, this approach is not that different than screening catalysts/ ligands to find out what works for a reaction. Groups have been doing that for years. I am wrong to think this is not that creative an idea?

Permalink to Comment

16. CMCguy on November 30, 2011 12:07 PM writes...

May be I am taking too literal but always believed Serendipity was more accidental/unexpected observations that provide novel results. Therefore not sure agree is a matter of simply probability and statistics and as others mention this is same trap combichem fell into for drug discovery. On the other hand this does fit well with Louis Pasteur's "Chance favors an open mind" so this would suggest a valid scientific approach. If work produces useful knowledge or better yet new transformation than can do things current chemistries can't or with improvements then may be of value.

Permalink to Comment

17. Phil on November 30, 2011 12:13 PM writes...

@16: You're absolutely right that this is nothing new. In fact, this kind of high-throughput experimentation (HTE) is already being used at Merck Process and other industrial labs to optimize reactions.

The cool part is the 1000 reactions per day. Hence, the "accelerated" part. Of course, as Derek points out (and what I know from my own experience), running the experiments is no longer the slow step. Analyzing the data becomes rate-limiting, which is fine. Get a computer to analyze the data for you and you go even faster.

Does this paper belong in Science (or Hartwig's for that matter)? I don't think so. The transformation in question isn't paradigm shifting, and the method of HTE is old news. Just because academics started investing money in it doesn't make it the new hotness.

Permalink to Comment

18. barry on November 30, 2011 12:51 PM writes...

there are more dangers than explosions in such an open-format experimentation. Back in Pittsburgh thirty-odd years ago professor Danishefsky nearly "lost" a grad student who had unwittingly synthesized a potent nerve gas.

Permalink to Comment

19. Anonymous on November 30, 2011 12:54 PM writes...

This would be Nobel material if the "computer/automation" could establish a new reaction product and optimize the yield without human intervention. Even better if CADD could also be enlisted in the predetermination of the substrate(s)/catalyst(s) process to narrow the field of possible reactions.
Let's see if the old way of optimization via numerous screening reactions by an exploited grad student will become obsolete in the future. I have always believed that it is a matter of time before the experimental organic chemist will become a relic as well!

Permalink to Comment

20. Curious Wavefunction on November 30, 2011 1:05 PM writes...

Anon: It's going to be a long time before CADD can measurably contribute to accurate prediction in this kind of situation. If you consider the myriad starting materials, products, metals, catalysts and solvent interactions, it basically translates to a many-body problem in which differences of a tenth of a kcal/mol or a hundredth of an angstrom in bond length can tip the scales. We are still far from being able to model the accurate solvation even of a simple arbitrary molecule in water. CADD could start contributing significantly at some point but for now, trial and error combined with semi-rational optimization is probably going to be more efficient than any computing power you throw at the problem.

Permalink to Comment