Artificial intelligence in small-molecule drug discovery

Over the last few years, the applications of artificial intelligence (AI) in drug discovery have grown substantially, raising the expectations about a near, radical AI-enabled transformation in drug discovery. The applications are becoming larger, from discovery of new molecules designed with neural networks to the in silico assessment of molecules under development. The promise of AI-enabled drug discovery and development is to greatly reduce costs and timelines for approval, as stated by several AI-native drug discovery companies. The interest in this topic is witnessed by the discovery partnerships that well-established pharmaceutical companies have formed with AI companies. As a paper recently published by a team of the Boston Consulting Group (BCG) explains (Tripathi et al., 2021), the dimensions for value creation for AI in drug discovery include, but are not limited to: - Greater productivity - Broader molecular diversity - Improved chances of clinical success Despite this progress, the level of adoption of these technologies is still limited and many open questions remain on their future impact. The present chapter discusses the wide range of AI applications in drug development, the pipeline progress of AI-enabled discovery programs, and the potential future outlook and challenges of this field.