>According to Grossman, the next step is “to use a combination of simulation, chemical intuition, and databases of tens of millions of known molecules to look for other candidates that have structural similarities and might exhibit the same behavior.”

That last bit was the most interesting part of the article to me. I've read and studied a fair amount about number crunching to try and figure out the behavior of proteins, but I haven't heard about anyone trying that in reverse and just throwing random chemicals in a simulation and seeing if they have the desired properties. It seems like that sort of chemical simulation for material science would predate folding at home and the like.