A multiobjective integer differential evolution approach for computer aided drug design

Computer aided drug discovery is a field where recent advances in technology are expected to impact in a significant way the traditional process of drug development. In this paper, an integer differential evolution (IDE) algorithm for De novo drug design is proposed. The aim is to find small molecules that are complementary in shape and charge to a reference molecule by making connections between fragments. The problem is cast as a multi-objective optimization task and the proposed IDE algorithm performs such task by aggregation of two objective functions namely Oral Bioavailability (OBA) and Tanimoto coefficient. An experimental study using libraries of acid and amine fragments as well as the Chemistry Development Kit (CDK) tools has been conducted and very promising results have been obtained. Better performance than a genetic algorithm with both objectives has been achieved.