Molegear: A Java-based platform for evolutionary de novo molecular design

Yunhan Chu, Xuezhong He*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

A Java-based platform, MoleGear, is developed for de novo molecular design based on the chemistry development kit (CDK) and other Java packages. MoleGear uses evolutionary algorithm (EA) to explore chemical space, and a suite of fragment-based operators of growing, crossover, and mutation for assembling novel molecules that can be scored by prediction of binding free energy or a weighted-sum multi-objective fitness function. The EA can be conducted in parallel over multiple nodes to support large-scale molecular optimizations. Some complementary utilities such as fragment library design, chemical space analysis, and graphical user interface are also integrated into MoleGear. The candidate molecules as inhibitors for the human immunodeficiency virus 1 (HIV-1) protease were designed by MoleGear, which validates the potential capability for de novo molecular design.

Original languageEnglish
Article number1444
JournalMolecules
Volume24
Issue number7
DOIs
StatePublished - 11 Apr 2019

Keywords

  • De novo design
  • Drug molecules
  • Evolutionary algorithm
  • Fitness
  • Multi-objective function

Fingerprint Dive into the research topics of 'Molegear: A Java-based platform for evolutionary de novo molecular design'. Together they form a unique fingerprint.

Cite this