Chasing the function of hypothetical proteins
Osnat Herzberg, Gary Gilliland, John Orban, Andrew Howard, John Moult

The focus of the Structure2Function program at CARB during the first five years was on prokaryotic proteins of unknown function, mostly from Heamophilus influenzae and E. coli. The goal was to leverage the 3-D structural information to gain insight into the biochemical function. Approximately 75% of the new structures exhibit folds that have been seen before, and in many cases we were able to make new discoveries about function by utilizing standard assays and mining genome context. The remaining 25% structures defined new folds, a somewhat higher fraction of new folds compared with structures entering the PDB. In some of these cases, the function became known around the time the structure was determined, while in other cases we were able to make broad predictions about the function.