Reverse Synthesis Pathway: Exploration of the Synthesis of o-Formylbenzoic Acid Methyl Ester Based on Deep Retrosynthesis Analysis

The significance of deep learning in designing compound synthesis routes is increasingly recognized. We have adopted a novel framework for deep retrosynthesis reaction prediction based on byproducts (RPBP). This framework leverages byproduct information to systematically formulate synthetic routes for target molecules, optimizing feasibility and enhancing synthesis efficiency. By integrating byproduct prediction with reactant generation, RPBP outperforms traditional methods in predictive accuracy and is applied here to design an efficient synthesis route for o-formylbenzoic acid methyl ester.