Kihara Lab Protein Docking Suite
The protein docking suite developed by our group includes programs to perform protein-protein docking prediction, multiple protein docking, as well as protein docking prediction using predicted protein-protein interfaces.
Our pairwise docking program, LZerD (Local 3D Zernike descriptor-based Docking algorithm), uses geometric hashing to generate ligand orientations using the 3D Zernike descriptors as the shape matching criteria. Scoring is based on a shape complementarity term defined by the local shape Zernike and orientation of the surface normals. In addition a clash penalty term and buried surface area have also been incorporated.
Multi-LZerD is a generic multiple protein docking program that can create docking models without additional biological information (like symmetry constraints). It is based on LZerD which uses geometric hashing and the 3D Zernike descriptor for pairwise docking purposes. Based on the pairwise predictions, Multi-LZerD applies a Genetic Algorithm to combine 2-body transformations in order to create a complete protein complex from the original parts. It uses a physics based score to guide the GA towards the best solution possible.
PI-LZerD uses protein interfaces predictions, from current protein interface prediction methods, in combination with LZerD to iteratively reach higher prediction accuracies.
|Copyright © 2017 KIHARA Bioinformatics LABORATORY, PURDUE University|