I just finished a piece for the March issue of Nature Chemistry on what (in my mind) make a molecule beautiful. I will admit a preference for sparer, less baroque structures. (If you want to know more about my molecular aesthetic, you'll have to wait for the piece to appear!). In the meantime there is an article in this month's Nature Chemistry with the intriguing title "Quantifying the Chemical Beauty of Drugs" [Bickerton et al. Nature Chem. 4, 93-97 (2012), full text is free]. It's not so much beauty in the abstract these chemists are trying to quantitatively capture, but desirability. How attractive is this molecule as a target for drug development? Would a chemist be willing to surrender time and bench space to the synthesis of this molecule?
The model takes as its inspiration Lipinski's rule of 5. If most or all of Lipinski's five characteristics are present, a molecule has a good chance of being a viable candidate for an oral drug [Lipinkski et al. Adv. Drug Dev. Rev. 23 3-25 (1997)]. The goal is to develop an expert model system, one that mimics (or improves on) a chemist's intuition about what makes for a good drug.
Earlier work had suggested that chemical fashion sense is drifting toward more baroque structures for their drugs, despite various rule sets that suggest that bloated molecules are less likely to survive to clinical trials. Chemists apparently like their molecules "tractable" (which would seem to mitigate against molecular overelaboration?), synthetically and otherwise! Molecular docility is desirable.
For a somewhat darker take on chemical intuition and seat of the pants drug design read "Chemists in the Shadows" by Adam Piore in March's Discover Magazine. The article focuses on underground chemists who are developing new recreational pharmaceuticals that skirt current drug laws (steroids for athletes, and rave drugs). The conceptual framework used by some of these chemists would be familiar to any medicinal chemist (particularly in the early days, before QSAR).
Kurt Gödel's Open World
1 day ago in The Curious Wavefunction