Illuminating elite patches of chemical space.

Journal: Chemical science
Published Date:

Abstract

In the past few years, there has been considerable activity in both academic and industrial research to develop innovative machine learning approaches to locate novel, high-performing molecules in chemical space. Here we describe a new and fundamentally different type of approach that provides a holistic overview of how high-performing molecules are distributed throughout a search space. Based on an open-source, graph-based implementation [J. H. Jensen, , 2019, , 3567-3572] of a traditional genetic algorithm for molecular optimisation, and influenced by state-of-the-art concepts from soft robot design [J. B. Mouret and J. Clune, , 2012, pp. 593-594], we provide an algorithm that (i) produces a large diversity of high-performing, yet qualitatively different molecules, (ii) illuminates the distribution of optimal solutions, and (iii) improves search efficiency compared to both machine learning and traditional genetic algorithm approaches.

Authors

  • Jonas Verhellen
    Centre for Integrative Neuroplasticity, University of Oslo N-0316 Oslo Norway jverhell@gmail.com.
  • Jeriek Van den Abeele
    Department of Physics, University of Oslo N-0316 Oslo Norway.

Keywords

No keywords available for this article.