Enhanced Chemical Reaction Mining through Combined Model Adaptation and Prompt Engineering
Journal of Chemical Information and Modeling · 2025
DOI: 10.1021/acs.jcim.5c02011 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, 2025. Verfasst von Madhavi Kumari und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pharmacoinformatics, National Institute of Pharmaceutical Education and Research (NIPER), Sector 67, S.A.S. Nagar (Mohali), Punjab 1600…. Hauptthemen laut MEDLINE: Data Mining, Cheminformatics. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 31 Literaturangaben. Gefördert von: Department of Biotechnology, Ministry of Science and Technology, India.
Abstract (Originalsprache)
Extracting structured chemical reaction information from the scientific literature remains challenging due to the complexity of chemical language and diverse reporting formats. This study presents a novel three-stage framework that combines quantized low-rank adaptation fine-tuning, entity self-grounding, and dynamic prompt engineering for enhanced chemical reaction extraction. The approach systematically addresses domain adaptation, accuracy refinement, and task optimization through progressive enhancement stages. Evaluation on 26,785 reaction descriptions demonstrates substantial improvements over existing methods, achieving F1 scores of 92.62% for reactants, 90.72% for products, 89.64% for solvents, and 88.83% for reagents. Chemical identifier recognition shows strong performance, with IUPAC names achieving 93.88% recall. The framework outperforms specialized systems like ChemRxnBERT (78.7% F1) and recent LLM approaches (83.0% F1) while requiring significantly less training data compared to domain-adaptive pretraining approaches and computational resources. This methodology establishes an effective framework for chemical reaction information extraction from unstructured text, enabling applications in automated literature mining, synthesis planning, and chemical informatics.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Pharmacoinformatics, National Institute of Pharmaceutical Education and Research (NIPER), Sector 67, S.A.S. Nagar (Mohali), Punjab 160062, India.
ORCID-Profile: Prabha Garg
Förderung
- Department of Biotechnology, Ministry of Science and Technology, India
Zitieren
APA
Kumari, M., & Garg, P. (2025). Enhanced Chemical Reaction Mining through Combined Model Adaptation and Prompt Engineering. Journal of Chemical Information and Modeling. https://doi.org/10.1021/acs.jcim.5c02011
Vancouver
Kumari M, Garg P. Enhanced Chemical Reaction Mining through Combined Model Adaptation and Prompt Engineering. Journal of Chemical Information and Modeling. 2025. doi:10.1021/acs.jcim.5c02011
BibTeX
@article{Kumari2025,
author = {Kumari, Madhavi and Garg, Prabha},
title = {Enhanced Chemical Reaction Mining through Combined Model Adaptation and Prompt Engineering},
journal = {Journal of Chemical Information and Modeling},
year = {2025},
doi = {10.1021/acs.jcim.5c02011}
}
EUVIMED