DEVELOPMENT AND APPLICATION OF STRUCTURE-ASSISTED ATOM-BASED DE NOVO MOLECULAR DESIGN (SAAD) APPROACH USING MATLAB

Abstract

In the context of drug design and development, phytochemical and synthetic libraries are starting places for searching leads. However, knowing crystal structure of receptor may help scratching molecules that complements available interaction sites using de novo molecular design approaches. This research describes the implementation of genetic algorithm optimization technique for construction of molecular scaffold composed from sp2 carbon, oxygen and nitrogen atoms. Using Matlab, functions for molecule construction, crossover, mutation and fitness measurement were individually programmed. Fitness function was derived from AutoDock3.05 function which calculates molecular interaction energy with protein using grid map. The approach was successfully applied to design glutathione-S-transferase inhibitory molecule at sub-site of the binding pocket that is originally occupied by g-glutamyl and cysteinyl moieties of glutathione substrate. Starting from scratches, the approach was able to design molecular scaffolds complementarily filling the pocket.