Graphical visualization of the main steps of the procedure used to identify the enzymatic cleavage site on peptide substrates.

a) Ten copies of the peptide substrate presenting a random conformation and position are generated. b) A large number of models of enzyme-substrate complexes are predicted during free docking simulations. The peptide ligand is fully flexible and samples the entire surface of the enzyme molecule. c) Hierarchical clustering is used to identify the most probable binding modes of a peptide substrate. Ten representative enzyme-substrate models are selected as central structures of the ten largest clusters identified in the entire model set. d) Histogram analysis of contacts between the substrate and the enzyme active site is used to identify the cleavage site using a large number of enzyme-protein complex models in all-atom representation. The figure shows the results of a single docking simulation of the substrate peptide (sequence: GAETF—YVDGA) to the HIV-1 protease. The enzyme protein is shown in blue surface representation. Peptides are shown in trace representation (peptide residues are colored from red to blue).