Scalable and Cost-Efficient de Novo Template-Based Molecular Generation

Piotr Gaiński, Oussama Boussif, Andrei Rekesh, Dmytro Shevchuk, Ali Parviz, Mike Tyers, Robert A. Batey, Michał Koziarski

Jagiellonian University· Mila· Université de Montréal· University of Toronto· The Hospital for Sick Children Research Institute· Acceleration Consortium· Vector Institute

drug discovery generative models GFlowNets synthesizability molecular generation

⋅ NeurIPS ⋅ Project Page ⋅OpenReview

Abstract

Template-based molecular generation offers a promising avenue for drug design by ensuring generated compounds are synthetically accessible through predefined reaction templates and building blocks.

In this work, we tackle three core challenges in template-based GFlowNets:

minimizing synthesis cost
scaling to large building block libraries
effectively utilizing small fragment sets.

We propose Recursive Cost Guidance, a backward policy framework that employs auxiliary machine learning models to approximate synthesis cost and viability. This guidance steers generation toward low-cost synthesis pathways, significantly enhancing cost-efficiency, molecular diversity, and quality, especially when paired with an Exploitation Penalty that balances the trade-off between exploration and exploitation.

To enhance performance in smaller building block libraries, we develop a Dynamic Library mechanism that reuses intermediate high-reward states to construct full synthesis trees. Our approach establishes state-of-the-art results in template-based molecular generation.