Ai Qianxiang, Meng Fanwang, Shi Jiale, Pelkie Brenden, Coley Connor W
Department of Chemical Engineering, Massachusetts Institute of Technology Cambridge MA USA
Department of Chemical Engineering, University of Washington Seattle WA USA.
Digit Discov. 2024 Jul 31;3(9):1822-1831. doi: 10.1039/d4dd00091a. eCollection 2024 Sep 11.
The popularity of data-driven approaches and machine learning (ML) techniques in the field of organic chemistry and its various subfields has increased the value of structured reaction data. Most data in chemistry is represented by unstructured text, and despite the vastness of the organic chemistry literature (papers, patents), manual conversion from unstructured text to structured data remains a largely manual endeavor. Software tools for this task would facilitate downstream applications such as reaction prediction and condition recommendation. In this study, we fine-tune a large language model (LLM) to extract reaction information from organic synthesis procedure text into structured data following the Open Reaction Database (ORD) schema, a comprehensive data structure designed for organic reactions. The fine-tuned model produces syntactically correct ORD records with an average accuracy of 91.25% for ORD "messages" (, full compound, workups, or condition definitions) and 92.25% for individual data fields (, compound identifiers, mass quantities), with the ability to recognize compound-referencing tokens and to infer reaction roles. We investigate its failure modes and evaluate performance on specific subtasks such as reaction role classification.
数据驱动方法和机器学习（ML）技术在有机化学及其各个子领域的普及提高了结构化反应数据的价值。化学领域的大多数数据都由非结构化文本表示，尽管有机化学文献（论文、专利）数量庞大，但从非结构化文本到结构化数据的手动转换在很大程度上仍然是一项人工工作。用于此任务的软件工具将促进下游应用，如反应预测和条件推荐。在本研究中，我们对一个大语言模型（LLM）进行微调，以按照开放反应数据库（ORD）模式将有机合成程序文本中的反应信息提取为结构化数据，ORD模式是一种为有机反应设计的综合数据结构。经过微调的模型生成语法正确的ORD记录，对于ORD“消息”（完整化合物、后处理或条件定义）的平均准确率为91.25%，对于单个数据字段（化合物标识符、质量数量）的平均准确率为92.25%，能够识别化合物引用令牌并推断反应角色。我们研究了其失败模式，并评估了在特定子任务（如反应角色分类）上的性能。
