Computer‐assisted synthetic planning considering reaction kinetics based on transition state automated generation method

Abstract Organic synthesis facilitates the conversion of raw materials into high‐value chemicals. Computer‐assisted synthetic planning plays a vital role in designing synthetic pathways, which are usually evaluated by the reaction probability using deep learning models. However, this criterion is generally hard to describe real reaction behaviors such as reaction kinetics. Therefore, this article aims to establish a reaction kinetics‐based retrosynthesis planning framework to design synthetic pathways with well‐performed reaction kinetics. The key contribution of this work is developing a method for the GENeration of initial guesses of Transition States based on Reactive Sites (GENiniTS‐RS) to automatically and fast generate the initial guesses of transition states for the transition state theory‐based reaction kinetic model without sampling the minimum energy path from reactants to products. Finally, two case studies involving the design of synthetic pathways for aspirin and ibuprofen are presented to demonstrate the feasibility and effectiveness of the proposed framework.