Leaders in Pharmaceutical Business Intelligence (LPBI) Group

ChatGPT Chemistry Assistant for Text Mining and the Prediction of MOF Synthesis

Reporter: Aviva Lev-Ari, PhD, RN

ChatGPT Chemistry Assistant for Text Mining and the Prediction of MOF Synthesis

• Zhiling Zheng
• Oufan Zhang
• Christian Borgs
• Jennifer T. Chayes
• Omar M. Yaghi*

Cite this: J. Am. Chem. Soc. 2023, 145, 32, 18048–18062 Publication Date:August 7, 2023 https://doi.org/10.1021/jacs.3c05819 Copyright © 2022 American Chemical Society. This publication is licensed under these Terms of Use.https://pubs.acs.org/doi/10.1021/jacs.3c05819

Abstract

We use prompt engineering to guide ChatGPT in the automation of text mining of metal–organic framework (MOF) synthesis conditions from diverse formats and styles of the scientific literature. This effectively mitigates ChatGPT’s tendency to hallucinate information, an issue that previously made the use of large language models (LLMs) in scientific fields challenging. Our approach involves the development of a workflow implementing three different processes for text mining, programmed by ChatGPT itself. All of them enable parsing, searching, filtering, classification, summarization, and data unification with different trade-offs among labor, speed, and accuracy. We deploy this system to extract 26 257 distinct synthesis parameters pertaining to approximately 800 MOFs sourced from peer-reviewed research articles. This process incorporates our ChemPrompt Engineering strategy to instruct ChatGPT in text mining, resulting in impressive precision, recall, and F1 scores of 90–99%. Furthermore, with the data set built by text mining, we constructed a machine-learning model with over 87% accuracy in predicting MOF experimental crystallization outcomes and preliminarily identifying important factors in MOF crystallization. We also developed a reliable data-grounded MOF chatbot to answer questions about chemical reactions and synthesis procedures. Given that the process of using ChatGPT reliably mines and tabulates diverse MOF synthesis information in a unified format while using only narrative language requiring no coding expertise, we anticipate that our ChatGPT Chemistry Assistant will be very useful across various other chemistry subdisciplines.

This publication is licensed for personal use by The American Chemical Society.

---

Our research has successfully demonstrated the potential of LLMs, particularly GPT models, in the domain of chemistry research. We presented a ChatGPT Chemistry Assistant that includes three different but connected approaches to text mining with ChemPrompt Engineering: Process 3 is capable of conducting search and filtration, Processes 2 and 3 classify synthesis paragraphs, and Processes 1, 2, and 3 are capable of summarizing synthesis conditions into structured data sets. Enhanced by three fundamental principles of prompt engineering specific to chemistry text processing, coupled with the interactive prompt refinement strategy, the ChatGPT-based assistant has substantially advanced the extraction and analysis of the MOF synthesis literature, with precision, recall, and F1 scores exceeding 90%.

We elucidated two crucial insights from the data set of synthesis conditions. First, the data can be employed to construct predictive models for reaction outcomes, which shed light on the key experimental factors that influence the MOF crystallization process. Second, it is possible to create an MOF chatbot that can provide accurate answers based on text mining, thereby improving access to the synthesis data set and achieving a data-to-dialogue transition. This investigation illustrates the potential for rapid advancement inherent in ChatGPT and other LLMs as a proof of concept.

On a fundamental level, this study provides guidance on interacting with LLMs to serve as AI assistants for chemists, accelerating research with minimal prerequisite coding expertise and thus bridging the gap between chemistry and the realms of computational and data science more effectively. Through interaction and chatting, the code and design of experiments can be modified, democratizing data mining and enhancing the landscape of scientific research. Our work sets a foundation for further exploration and application of LLMs across various scientific domains, paving the way for a new era of AI-assisted chemistry research.

These datasets on the synergy of the highly-porous materials known as metal-organic frameworks (MOFs) will inform predictive models. The models will accelerate chemists’ ability to create or optimize MOFs, including ones that alleviate water scarcity and capture air pollution. All chemists – not just coders – can build these databases due to the use of AI-fueled chatbots.

To help them teach ChatGPT to generate accurate and relevant information, they modified an approach called “prompt engineering” into “ChemPrompt Engineering.” They developed prompts that avoided asking ChatGPT for made up or misleading content; laid out detailed directions that explained to the chatbot the context and format for the response; and provided the large language model a template or instructions for extracting data.

The chatbot’s literature review – and the experts’ approach – was successful. ChatGPT finished in a fraction of an hour what would have taken a student years to complete, said Borgs, BIDMaP’s director. It mined the synthetic conditions of MOFs with 95% accuracy, Yaghi said.