A Graph-Based Genetic Algorithm and Its Application to the Multiobjective Evolution of Median Molecules

J Chem Inf Comput Sci. May-Jun 2004;44(3):1079-87. doi: 10.1021/ci034290p.

Abstract

In this paper we propose a novel graph-based genetic algorithm for the evolution of novel molecular graphs from a predefined set of elements or molecular fragments with an external objective function. A brief overview of existing genetic algorithm approaches in molecular design is provided followed by a description of our approach. The paper continues to suggest a novel application of this program to the multiobjective evolution of median molecules that are structurally representative of a set of objective molecules. We conclude with a summary of our initial results along with a discussion of a variety of improvements and applications of our approach.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms*
  • Biological Evolution*
  • Computer-Aided Design
  • Mutation