Drug discovery with explainable artificial intelligence
It tackles the problem of opaque AI models in molecular sciences, but is incremental as it reviews existing methods without presenting new results.
This review addresses the lack of interpretability in deep learning models used for drug discovery, summarizing key explainable AI concepts and forecasting future opportunities and challenges.
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.