Robot-mediated development of statistical models for mechanistic analysis amplification in synthetic organic reactions

Cyb-org aims to apply well-established machine learning (ML) techniques to innovate mechanistic studies and optimization protocols in organic chemistry. Within this project, a robotic platform will be programmed to autonomously set up and sample reactions for kinetic analysis, extracting data on reaction rates, yields, and enantioselectivity. The primary goal of this proposal is to develop statistical models capable of generalizing mechanistic insights across a range of substrates or catalyst variables, focusing on the selected organic transformations. The new model will generate predicted outputs, which can subsequently be experimentally validated to achieve enhanced performance. Identifying outliers within this data will allow for the analysis of atypical reactions, potentially leading to the discovery of novel reactivity and the development of new, significant synthetic strategies. This advanced ML tool could be adopted by the pharmaceutical and fine chemical industries, enhancing automated systems for optimizing organic chemical transformations.