Over at In the Pipeline, Derek has posted a great statement about The Hazards of Molecular Modeling:
I’ve made this point before, but it needs to be made again: molecular modeling is not reality. Most models are not that good, or only good around a limited group of rather similar compounds. If you as a medicinal chemist are crossing out easy-to-make compounds in unexplored chemical space just because the software doesn’t like it, you are handcuffing yourself and tying your thumbs together. Stop it, stop it for your own good, or you may never discover anything unexpected or useful.
As someone who is in the business of developing computational methods for molecular modeling, I have this to say: RIGHT ON!
I can’t emphasize enough how far from reality some of our force fields actually are. There are only a few force fields with coupling terms, a vanishingly small number that can actually break and reform bonds, and no quantum mechanics at all. The problem is that the computational tools have gotten so easy to use that the original developers of these tools (who usually understand the limits of their models) aren’t the ones using the codes. In the hands of non-experts, these tools can present an incorrect version of reality. When I teach Computational Chemistry, I try to instill a healthy skepticism about every single topic we cover.