AI has garnered much interest and significant investment dollars in recent years but has yet to give a step change in the new molecule design process. In contrast, techniques like free energy perturbation (FEP) have enabled significant digitization of previously wet chemistry-led workflows but these techniques remain costly to run at scale.

In this paper we will discuss the advantages of AI, ML, and FEP and how combining them promises a paradigm shift in a medicinal chemist’s ability to progress molecules from hit to lead and on to the drug candidate in silico, reducing time and spend in this critical part of the process.