Using Artificial Intelligence forde novoDrug Design and Retrosynthesis
Rohit Romesh Arora, Nicolas Brosse, Clarisse Descamps, Nicolas Devaux, Nicolas Do Huu, Philippe Gendreau, Yann Gaston‐Mathé, Maud Parrot, Quentin Perron, Hamza Tajmouati · 2024
Artificial intelligence (AI), driven by progress in deep learning, has emerged as a disruptive technology across multiple industries. The development of AI approaches to de novo drug design through the use of deep generative models has triggered significant interest in the drug-hunter community as an important tool to accelerate the process. These approaches have attracted significant interest as they are seen to have advantages not present in classic expert-based technologies, especially in addressing challenges such as multiparameter optimization (MPO). In this chapter, we explore the various AI algorithms and modes that are being deployed to accelerate and address medicinal chemistry challenges across the drug discovery value chain – from hit discovery, to hit-to-lead, to lead optimization. We especially explore the importance of integrating synthetic accessibility to generative algorithms and its impact on the quality of novel designed compounds. These technologies have the potential to transform the drug discovery landscape in the coming decade.