Scalable and Cost-Efficient de Novo Template-Based Molecular Generation
This work addresses cost and scalability issues in drug design, offering incremental improvements to template-based GFlowNets for generating synthetically accessible molecules.
The paper tackled challenges in template-based molecular generation for drug design, specifically minimizing synthesis cost, scaling to large building block libraries, and utilizing small fragment sets, by proposing Recursive Cost Guidance and a Dynamic Library mechanism, achieving state-of-the-art results.
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: (1) minimizing synthesis cost, (2) scaling to large building block libraries, and (3) 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.