Exploring Spectrum‐based Molecular Descriptors for Reaction Performance Prediction

• Published in: Chemistry, an Asian journal Vol. 18; no. 7; pp. e202300011 - n/a
• Main Authors: Tang, Miao‐Jiong, Xu, Li‐Cheng, Zhang, Shuo‐Qing, Hong, Xin
• Format: Journal Article
• Language: English
• Published: Germany Wiley Subscription Services, Inc 03.04.2023
• Subjects: Chemistry; Cross coupling; Halides; Machine learning; molecular descriptor; Optical character recognition; Performance prediction; quantitative structure-activity relationship; spectrum; yield prediction; molecular descriptor; yield prediction; quantitative structure-activity relationship; machine learning; spectrum
• ISSN: 1861-4728; 1861-471X; 1861-471X
• DOI: 10.1002/asia.202300011

Despite the availability and accuracy of modern spectroscopic characterization, the utilization of spectral information in chemical machine learning is still primitive. Here, we report an optical character recognition‐based automatic process to utilize spectral information as molecular descriptors, which directly transforms experimental spectrum images to readable vectors. We demonstrate its machine learning application in the reaction yield dataset of Pd‐catalyzed Buchwald‐Hartwig cross‐coupling with aryl halides. In addition, we also show that the predicted spectrum can serve as an alternative encoding source to support the model training. Spectroscopy, as one of the most widely applied characterization techniques, contains a wealth of chemical information, yet its application in chemical machine learning is still limited. This work reports the OCR processing of spectrum images to machine learning‐readable descriptors, and these descriptors were found effective in the modeling of yield prediction.