On the way to real time protein-ligand sampling

Protein-ligand binding free energy is one of the keystones of drug design, and developing a fast method to calculate it would have great impact in personalized medicine. However, it is a daunting task for computational methods, since the conformational space is rugged, having a lot of metastable states that hinder the exploration. Using PELE and an adaptive sampling scheme, one can quickly get thermodynamic properties by traversing the conformational space on a simulation time scale (24h). We show the performance on a new benchmark of a series of different families of proteins and ligands with a large range of binding free energy differences (about 8 kcal/mol).