EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SynthStrategy: Extracting and Formalizing Latent Strategic Insights from LLMs in Organic Chemistry

Daniel Armstrong iD, Zlatko Jončev, Andres M Bran, Xuan Vu Nguyen, Théo A Neukomm iD, Philippe Schwaller iD

Journal of Chemical Information and Modeling · 11.11.2026

DOI: 10.1021/acs.jcim.6c01688 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, online erschienen am 11.11.2026. Verfasst von Daniel Armstrong und 5 weiteren Autor:innen. Beteiligte Einrichtung: Laboratory of Artificial Chemical Intelligence (LIAC), Institute of Chemical Sciences and Engineering (ISIC), École Polytechnique Fédérale de Lausann…. Die Arbeit nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCCR Catalysis, Swiss National Science Foundation (SNSF).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern computer-assisted synthesis planning (CASP) systems show promise 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 (LLMs) to distill synthetic knowledge into code. Our system analyzes synthesis routes and translates strategic principles into executable Python functions representing diverse strategic and tactical rules, such as strategic functional group interconversions, protecting strategies and ring construction features. We validate the use of LLM-as-a-judge in this domain, demonstrating their efficacy in evaluating synthesis-specific code generation. 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 data set─synthesis routes annotated with strategic tags. This framework unlocks a novel capability for CASP: natural language-based route retrieval, achieving 74.5% 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.

Autor:innen und Institutionen

  • Laboratory of Artificial Chemical Intelligence (LIAC), Institute of Chemical Sciences and Engineering (ISIC), École Polytechnique Fédérale de Lausanne (EPFL), Station 6, LausanneCH-1015, Switzerland.
  • b-12 Labs, San Francisco, California94107, United States.
  • National Centre of Competence in Research (NCCR) Catalysis, École Polytechnique Fédérale de Lausanne (EPFL), LausanneCH-1015, Switzerland.

ORCID-Profile: Daniel Armstrong, Théo A Neukomm, Philippe Schwaller

Förderung

  • NCCR Catalysis
  • Swiss National Science Foundation (SNSF)

Zitieren

APA

Armstrong, D., Jončev, Z., Bran, A. M., Nguyen, X. V., Neukomm, T. A., & Schwaller, P. (2026). SynthStrategy: Extracting and Formalizing Latent Strategic Insights from LLMs in Organic Chemistry. Journal of Chemical Information and Modeling. https://doi.org/10.1021/acs.jcim.6c01688

Vancouver

Armstrong D, Jončev Z, Bran AM, Nguyen XV, Neukomm TA, Schwaller P. SynthStrategy: Extracting and Formalizing Latent Strategic Insights from LLMs in Organic Chemistry. Journal of Chemical Information and Modeling. 2026. doi:10.1021/acs.jcim.6c01688

BibTeX
@article{Armstrong2026,
  author = {Armstrong, Daniel and Jončev, Zlatko and Bran, Andres M and Nguyen, Xuan Vu and Neukomm, Théo A and Schwaller, Philippe},
  title = {SynthStrategy: Extracting and Formalizing Latent Strategic Insights from LLMs in Organic Chemistry},
  journal = {Journal of Chemical Information and Modeling},
  year = {2026},
  doi = {10.1021/acs.jcim.6c01688}
}
Kostenlos starten →