Artificial Intelligence in Drug Discovery: From Computational Target Identification to Clinical Translation

Abstract

Artificial intelligence (AI) has become an integral component of modern drug discovery, emerging as a structured and data-driven approach to address long-standing challenges such as rising development costs, extended timelines, and high failure rates. Traditional drug development processes remain inefficient, with many candidates failing in late-stage clinical trials due to inadequate efficacy, safety concerns, or unfavourable pharmacokinetics. Recent progress in computational capabilities, coupled with the rapid growth of large-scale biomedical and clinical datasets, has enabled broader and more reliable application of AI methodologies, including machine learning (ML), deep learning (DL), and natural language processing (NLP), across the drug discovery pipeline. These approaches support the integration and analysis of complex, high-dimensional data generated from genomics, proteomics, transcriptomics, chemical libraries, and real-world clinical sources. This review outlines the application of artificial intelligence (AI) across the drug discovery pipeline, from target identification and lead optimisation to preclinical evaluation and clinical translation, with emphasis on virtual screening, de novo design, and pharmacokinetic and toxicity prediction.