Graph Energy-based Model for Substructure Preserving Molecular Design
This method provides a new way for chemists to incorporate their domain knowledge into molecular design by enabling substructure-preserving generation.
The paper addresses the challenge of generating novel molecules while preserving a target substructure, a common practice for chemists. Their Graph Energy-based Model (GEM) successfully generates novel molecules while maintaining specified substructures, as demonstrated on chemistry datasets.
It is common practice for chemists to search chemical databases based on substructures of compounds for finding molecules with desired properties. The purpose of de novo molecular generation is to generate instead of search. Existing machine learning based molecular design methods have no or limited ability in generating novel molecules that preserves a target substructure. Our Graph Energy-based Model, or GEM, can fix substructures and generate the rest. The experimental results show that the GEMs trained from chemistry datasets successfully generate novel molecules while preserving the target substructures. This method would provide a new way of incorporating the domain knowledge of chemists in molecular design.