SynthStrategy: Extracting and Formalizing Latent Strategic Insights from LLMs in Organic Chemistry
This work addresses the tactical-strategic divide in organic chemistry synthesis planning, enabling specification, search, and evaluation of routes by strategic criteria rather than structure alone.
The paper tackles the problem of computer-assisted synthesis planning systems lacking strategic considerations like convergent assembly and protecting group minimization by leveraging Large Language Models to distill synthetic knowledge into verifiable code, achieving 75% Top-3 accuracy on a benchmark for natural language-based route retrieval.
Modern computer-assisted synthesis planning (CASP) systems show promises at generating chemically valid reaction steps but struggle to incorporate strategic considerations such as convergent assembly, protecting group minimization, and optimal ring-forming sequences. We introduce a methodology that leverages Large Language Models to distill synthetic knowledge into code. Our system analyzes synthesis routes and translates strategic principles into Python functions representing diverse strategic and tactical rules, such as strategic functional group interconversions and ring construction strategies. By formalizing this knowledge as verifiable code rather than simple heuristics, we create testable, interpretable representations of synthetic strategy. We release the complete codebase and the USPTO-ST dataset -- synthesis routes annotated with strategic tags. This framework unlocks a novel capability for CASP: natural language-based route retrieval, achieving 75\% Top-3 accuracy on our benchmark. We further validate our library through temporal analysis of historical trends and chemically intuitive route clustering that offers more granular partitioning than common previous methods. This work bridges the tactical-strategic divide in CASP, enabling specification, search, and evaluation of routes by strategic criteria rather than structure alone.