Actual and potential subsites are mapped onto the representative 3D structures from four families in the validation dataset. Actual subsites are shown in space fill model and colored in red while subsites that were commonly predicted by all three methods (C3 sites) or by any pair of methods (C2 sites) are shown in stick and colored in green and blue, respectively. Specific ligands and/or cofactors are shown in purple and cyan. a) IDH_IMDH family (1AI2; ligand: isocitrate; cofactor: NADP) b) nucleotidyl cyclase family (1CS4, chain A; activator: forskolin; P-site inhibitor: 2'-d-3'-AMP.PPi) c) serine protease family (5PTP; ligand: specificity determining serine residue at position 195) and d) LDH_MDH family (9LDT, chain A; ligand: oxamate; cofactor: NAD). The figure was generated using the PyMOL software .