Fig. 1

Visualization of ligand binding sites predicted by P2Rank for structure 1FBL. Protein is covered in a layer of points lying on the Solvent Accessible Surface of the protein. Each point represents its local chemical neighborhood and is colored according to its predicted ligandability score (from 0 = green to 1 = red). Points with high ligandablity score are clustered to form predicted binding sites (marked by coloring adjacent protein surface). In this case, the largest predicted pocket (shown in the close-up) is indeed a correctly predicted true binding site that binds a known ligand (magenta). Visualization is based on a PyMOL script produced by P2Rank