Large-scale comparison of machine learning methods for drug target prediction on ChEMBL

TLDR

Deep learning has emerged as a powerful tool in drug discovery, yet its superiority over traditional computational methods remains uncertain due to dataset bias and hyperparameter selection challenges. The study aimed to evaluate the performance of multiple deep learning models on a large drug discovery dataset and benchmark them against other machine learning and target‑prediction approaches. To mitigate bias, the authors employed a nested cluster‑cross‑validation strategy that controls for compound series and hyperparameter selection. Results showed that deep learning models consistently outperformed all competing methods and, in many cases, matched the predictive accuracy of in‑vitro laboratory assays.

Abstract

Deep learning is currently the most successful machine learning technique in a wide range of application areas and has recently been applied successfully in drug discovery research to predict potential drug targets and to screen for active molecules. However, due to (1) the lack of large-scale studies, (2) the compound series bias that is characteristic of drug discovery datasets and (3) the hyperparameter selection bias that comes with the high number of potential deep learning architectures, it remains unclear whether deep learning can indeed outperform existing computational methods in drug discovery tasks. We therefore assessed the performance of several deep learning methods on a large-scale drug discovery dataset and compared the results with those of other machine learning and target prediction methods. To avoid potential biases from hyperparameter selection or compound series, we used a nested cluster-cross-validation strategy. We found (1) that deep learning methods significantly outperform all competing methods and (2) that the predictive performance of deep learning is in many cases comparable to that of tests performed in wet labs (i.e., in vitro assays).