Structure-based drug discovery with deep learning
It highlights a shift towards AI-guided structure-based methods in drug discovery, which could improve efficiency in pharmaceutical research, but is incremental as it reviews existing concepts rather than introducing new breakthroughs.
This review addresses the potential of deep learning to tackle unsolved challenges in structure-based drug discovery, such as affinity prediction for unexplored protein targets, by summarizing prominent algorithmic concepts and forecasting future opportunities.
Artificial intelligence (AI) in the form of deep learning bears promise for drug discovery and chemical biology, $\textit{e.g.}$, to predict protein structure and molecular bioactivity, plan organic synthesis, and design molecules $\textit{de novo}$. While most of the deep learning efforts in drug discovery have focused on ligand-based approaches, structure-based drug discovery has the potential to tackle unsolved challenges, such as affinity prediction for unexplored protein targets, binding-mechanism elucidation, and the rationalization of related chemical kinetic properties. Advances in deep learning methodologies and the availability of accurate predictions for protein tertiary structure advocate for a $\textit{renaissance}$ in structure-based approaches for drug discovery guided by AI. This review summarizes the most prominent algorithmic concepts in structure-based deep learning for drug discovery, and forecasts opportunities, applications, and challenges ahead.