Drug discovery with explainable artificial intelligence

TL;DR

This paper reviews how explainable AI can enhance drug discovery by making deep learning models more interpretable, addressing the need for transparency in molecular science applications.

Contribution

It summarizes key explainable AI algorithms and discusses future opportunities, applications, and challenges in applying these methods to drug discovery.

Findings

Summarizes prominent explainable AI algorithms for drug discovery.

Forecasts future applications and challenges in the field.

Highlights the importance of interpretability in molecular AI models.

Abstract

Deep learning bears promise for drug discovery, including advanced image analysis, prediction of molecular structure and function, and automated generation of innovative chemical entities with bespoke properties. Despite the growing number of successful prospective applications, the underlying mathematical models often remain elusive to interpretation by the human mind. There is a demand for 'explainable' deep learning methods to address the need for a new narrative of the machine language of the molecular sciences. This review summarizes the most prominent algorithmic concepts of explainable artificial intelligence, and dares a forecast of the future opportunities, potential applications, and remaining challenges.