From machine learning to deep learning: Advances in scoring functions for protein–ligand docking

TLDR

Molecule docking is a routine tool for drug discovery, but its accuracy depends on reliable scoring functions; machine learning–based scoring functions have emerged as promising alternatives with better performance than classical methods, and deep learning approaches are now being explored for even greater accuracy. We aim to summarize the progress of traditional ML‑based scoring functions and provide insights into recently developed DL‑based scoring functions. We review recent developments in ML‑ and DL‑based scoring functions for protein–ligand docking. We believe that continuous improvement in ML‑based scoring functions can guide early‑stage drug design and accelerate new drug discovery. The article is categorized under Computer and Information Science > Chemoinformatics.

Abstract

Abstract Molecule docking has been regarded as a routine tool for drug discovery, but its accuracy highly depends on the reliability of scoring functions (SFs). With the rapid development of machine learning (ML) techniques, ML‐based SFs have gradually emerged as a promising alternative for protein–ligand binding affinity prediction and virtual screening, and most of them have shown significantly better performance than a wide range of classical SFs. Emergence of more data‐hungry deep learning (DL) approaches in recent years further fascinates the exploitation of more accurate SFs. Here, we summarize the progress of traditional ML‐based SFs in the last few years and provide insights into recently developed DL‐based SFs. We believe that the continuous improvement in ML‐based SFs can surely guide the early‐stage drug design and accelerate the discovery of new drugs. This article is categorized under: Computer and Information Science > Chemoinformatics