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: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
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
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
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
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
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

« Remember When We Were Going to Eliminate Deaths from Cancer? | Main | Independent Inventions »

February 23, 2012

The Worst Compound You've Ever Drawn?

Email This Entry

Posted by Derek

Stuart Cantrill has a post on one of those vast dendrimer structures - you know, those mandala-like things that weigh as much as a beer truck. He says that if you can draw the structure on his page in ChemDraw (or the like) in under three hours, you are clearly a wonder-worker.

He's asking on his Twitter feed for examples of the worst chemical structure anyone's had to draw, so I thought I'd throw the same question out to the crowd. You're going to have had to have lead an evil past life to be able to beat his dendrimer, though.

Comments (17) + TrackBacks (0) | Category: Chemical News


1. Hap on February 23, 2012 1:29 PM writes...

A paper of Sharpless's nonracemic dendrimers - they weren't as big but the branching was harder (with stereo, etc.). To make it worse, I had to draw them twice. Yuck.

The natural product rugulosin (KCN made it awhile ago) is also a pain - it's not big but it's so dense with its polycyclic core that it's hard to draw in any way that doesn't either look like crap or be completely uninterpretable.

Permalink to Comment

3. paperclip on February 23, 2012 1:41 PM writes...

This is a self-imposed evil: For a proposal in grad school I drew dendrimers with Taxol drugs appended at the ends. (I may be a masochist.)

Some of the siderophores are a pain to draw, too, like ferrocrocin.

Permalink to Comment

4. Curryworks on February 23, 2012 1:52 PM writes...

At least they are honest down under. Academics pretending that academic integrity is king get that shoved in their face. The odd part is that it is simply "publish or perish" being put to work who knew. Now at every lecture one attends on a topic slightly related to academic reform this anecdote will be shared.

Permalink to Comment

5. RM on February 23, 2012 2:08 PM writes...

The compounds that I hate to draw are the ones that aren't "flat".

Heavily substituted chair-conformer sugars in Haworth-type projection are a persistent annoyance (especially when ChemDraw insists on flipping the orientations of the, e.g. -OCH3's to the "wrong" way), but there are other ones which are a pain, especially with natural products. Things like the opioids come to mind, although their simple enough that if you do it once, you can copy-paste and not have to worry about it again.

It seems like it's mainly natural products which have this non-flat issue. Synthetic drugs tend to be much more amenable to a flat representation. I wonder why that is. It might be that current synthetic procedure is more likely to come up with simple, "flat" structures, but I worry a bit that having to draw a synthetic scheme might subtly constrain chemists - unconsciously rejecting working in the third dimension because "if you can't draw it, it must be impossible".

Permalink to Comment

6. 6. on February 23, 2012 4:14 PM writes...

I feel bad for the KC Nicolaou student who has to give a chalk talk on maitotoxin.

Permalink to Comment

7. Anonymous on February 23, 2012 10:35 PM writes...

Organometallics can get awfully nasty. Inevitable one must compromise on the triphenylphosphenes and make some bonds really long just to fit the other five substituents in the same vicinity of the metal they are all attached to. How many substituents can you put onto one center, all of the while the carbon bit has to match all of the other carbon bits in the paper.

Permalink to Comment

8. scientistbymistake on February 24, 2012 5:54 AM writes...

I had to draw a few mycolic acids for my PhD thesis, count those carbons!
A few chiral-cyclopropane containing sponge products gave me eye-strain as well, but neither of these compete with the aforementioned dendrimers.

Permalink to Comment

9. anon the ii on February 24, 2012 7:58 AM writes...

Enough of the whining. Having drawn structures for my thesis with rub-ons, templates and a rapid-o-graph, I enjoy using ChemDraw to draw structures, the more complicated, the better. And figuring out how to make a complex, non-flat natural product convey both shape and connectivity at the same time is a rewarding challenge. This is the fun stuff. I think RM is in the wrong business.
If you want to talk about misery, let's talk about emulsions.

Permalink to Comment

10. scientistbymistake on February 24, 2012 8:58 AM writes...

@no. 10
Well that's us told!

Permalink to Comment

11. milo on February 24, 2012 9:52 AM writes...


Emulsions are easy to draw.

Permalink to Comment

12. InfMP on February 24, 2012 11:42 AM writes...

Worst drawings of the simplest compounds:


Permalink to Comment

13. DRZ on February 24, 2012 3:57 PM writes...

There's software that can understand a structural formula in an image file and output a SMILES string that you can paste into Chemdraw or other programs.

For a free program just search OSRA: Optical Structure Recognition. Follow the link for the web interface. Open a high resolution png or jpg of a molecule (wikipedia is great source) hit submit and it will spit out the SMILES or an sd file. It might take a minute or two with large structures. Very easy to try.

With complicated structures you'll probably have to do some corrections, but it can still be pretty helpful. It does an OK job on something like Vitamin B-12, for instance, which would probably take 20 minutes to draw on your own.

I actually made a short tutorial on this but haven't posted it anywhere.

Tried it on the dendrimer. Didn't work. Resolution of the image might not be good enough.

Permalink to Comment

14. midwestern on February 24, 2012 9:47 PM writes...

Edge-on views of porphyrins with a hexa-coordinate metal center. Ugh.

Permalink to Comment

15. Anonymous on February 24, 2012 10:10 PM writes...

If I ever had to draw a carbon nanotube with a specific chirality, I think I would cry.

Permalink to Comment

16. Anonymous on February 25, 2012 8:22 AM writes...

I was a post-doc in Fréchet's group and had a LOT of fun drawing dendrimers with ChemDraw.

But beyond the fun, we should not miss an essential point: drawing structures can actually give you ideas on how to synthesize the actual molecule. Especially if they're complex to draw.

Dendrimers are a perfect example: at the beginning, I started from the core, drawing them outward. Well, it's tough: angles are easy to define because of symmetry, but bond length is a critical parameter, and you can't foresee the length you need to adopt to make the drawing nice and symmetrical. I failed a number of times to get to the generation number I needed to get to, and the real problem is that you can't tweak the final drawing, you need to start all over.
Then I discovered the trick : don't start from the core, start from the end group. Draw one. Copy it. Align both. Rotate one (you can easily calculate the angle you need if you know the number of end groups that must span 360 degrees). Join them through the branched monomer. Copy the whole thing. Rotate. Join them through the branched monomer. Copy the whole thing. etc. Simple, and it works all the time, for any dendrimer.

Now, what you get here is the difference between the divergent synthesis of dendrimers (Tomalia et al.), where the number of reaction points increases exponentially with generation, and the convergent synthesis, where you only have two reaction points, at each generation (Fréchet et al.). What difference does it make?
In the divergent approach, at generation (="layer") 3, for example, you have 8 reaction points. The probability that you miss at least one of them (in the flask) is high and (most importantly), the perfect dendrimer and the defective dendrimer are two objects that look very much alike: they're going to be difficult to separate from one another. And this is only generation 3: at generation 4, you have 16 reaction points, and the perfect and defective dendrimers are even more similar.
In the convergent approach, since you only have 2 reactions points, the probability that you miss one of them is low, and even if you do, the perfect and defective dendrimers are very different from one another (the defective dendrimer is about half the size of the perfect one) : they are easy to separate.
All in all, with the convergent approach, you're able to build high-molecular weight (5000+ g.mol-1) molecules with the same precision as any other small molecule. This monodispersity allows you to make precise (and reliable) measurements on the properties of these macromolecules.

Conclusion: trying to draw the dendrimer can give you the idea of the convergent synthesis. I'm not sure that's how Fréchet got the idea in the first place, but it could have been.

Another example: sometimes, when you try to draw "complex" natural products, you realize that some parts are very similar and that you can copy-paste them: it could indeed come from dimerization... now that may be the start of a new synthetic route!

Permalink to Comment

17. newnickname on February 27, 2012 2:17 PM writes...

@15: Nanotube Modeler by jcrystal . com is or was free. It can draw any (n,m) SWNT, cone, sheet or ball. The latest version appears to be 1.7.3. It works for me.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry