Directly Optimizing for Synthesizability in Generative Molecular Design using Retrosynthesis Models

TL;DR

This paper demonstrates that with a sample-efficient generative model, it is possible to directly optimize for synthesizability in molecular design using retrosynthesis models within a computationally constrained environment.

Contribution

The authors introduce a method to integrate retrosynthesis models directly into the generative process for synthesizability optimization, overcoming previous inference cost limitations.

Findings

Generated molecules meet drug discovery criteria and are synthesizable according to retrosynthesis.

The approach operates effectively under limited computational resources.

It enables goal-directed molecular generation with synthesizability constraints.

Abstract

Synthesizability in generative molecular design remains a pressing challenge. Existing methods to assess synthesizability span heuristics-based methods, retrosynthesis models, and synthesizability-constrained molecular generation. The latter has become increasingly prevalent and proceeds by defining a set of permitted actions a model can take when generating molecules, such that all generations are anchored in "synthetically-feasible" chemical transformations. To date, retrosynthesis models have been mostly used as a post-hoc filtering tool as their inference cost remains prohibitive to use directly in an optimization loop. In this work, we show that with a sufficiently sample-efficient generative model, it is straightforward to directly optimize for synthesizability using retrosynthesis models in goal-directed generation. Under a heavily-constrained computational budget, our model can generate molecules satisfying a multi-parameter drug discovery optimization task while being synthesizable, as deemed by the retrosynthesis model.