Synthelite: Chemist-aligned and feasibility-aware synthesis planning with LLMs

TL;DR

Synthelite leverages large language models to create flexible, expert-interactive synthesis planning routes that consider chemical feasibility, significantly advancing computer-aided synthesis planning by integrating human insights.

Contribution

Introduces Synthelite, a novel LLM-based synthesis planning framework enabling expert interaction and feasibility-aware route generation, addressing limitations of previous CASP systems.

Findings

Achieves up to 95% success rate in constrained synthesis tasks

Allows natural language expert intervention during planning

Accounts for chemical feasibility in route design

Abstract

Computer-aided synthesis planning (CASP) has long been envisioned as a complementary tool for synthetic chemists. However, existing frameworks often lack mechanisms to allow interaction with human experts, limiting their ability to integrate chemists' insights. In this work, we introduce Synthelite, a synthesis planning framework that uses large language models (LLMs) to directly propose retrosynthetic transformations. Synthelite can generate end-to-end synthesis routes by harnessing the intrinsic chemical knowledge and reasoning capabilities of LLMs, while allowing expert intervention through natural language prompts. Our experiments demonstrate that Synthelite can flexibly adapt its planning trajectory to diverse user-specified constraints, achieving up to 95\% success rates in both strategy-constrained and starting-material-constrained synthesis tasks. Additionally, Synthelite exhibits the ability to account for chemical feasibility during route design. We envision Synthelite to be both a useful tool and a step toward a paradigm where LLMs are the central orchestrators of synthesis planning.