Mind the Retrosynthesis Gap: Bridging the divide between Single-step and Multi-step Retrosynthesis Prediction

TL;DR

This paper investigates how single-step retrosynthesis models can be effectively transferred to multi-step synthesis planning, demonstrating significant performance improvements and highlighting the need for joint evaluation metrics.

Contribution

It establishes a bridge between single-step and multi-step retrosynthesis, benchmarking transferability, and emphasizing the importance of multi-step evaluation metrics.

Findings

Single-step models can improve multi-step route finding by up to 30%.

No clear correlation exists between single-step and multi-step evaluation metrics.

Developing models specifically for multi-step retrosynthesis is necessary.

Abstract

Retrosynthesis is the task of breaking down a chemical compound recursively step-by-step into molecular precursors until a set of commercially available molecules is found. Consequently, the goal is to provide a valid synthesis route for a molecule. As more single-step models develop, we see increasing accuracy in the prediction of molecular disconnections, potentially improving the creation of synthetic paths. Multi-step approaches repeatedly apply the chemical information stored in single-step retrosynthesis models. However, this connection is not reflected in contemporary research, fixing either the single-step model or the multi-step algorithm in the process. In this work, we establish a bridge between both tasks by benchmarking the performance and transfer of different single-step retrosynthesis models to the multi-step domain by leveraging two common search algorithms, Monte Carlo Tree Search and Retro*. We show that models designed for single-step retrosynthesis, when extended to multi-step, can have a tremendous impact on the route finding capabilities of current multi-step methods, improving performance by up to +30% compared to the most widely used model. Furthermore, we observe no clear link between contemporary single-step and multi-step evaluation metrics, showing that single-step models need to be developed and tested for the multi-step domain and not as an isolated task to find synthesis routes for molecules of interest.