Drug discovery is a multifaceted process that encompasses several critical stages, including target identification, compound screening, lead optimization, preclinical studies, and clinical evaluation (Sun et al., 2022). It is widely recognized that drug development is often time-consuming and expensive (Hinkson et al., 2020). Approximately 90 % of candidates fail during clinical trials, often due to inadequate target selection (Dowden and Munro, 2019; Takebe et al., 2018). Early and accurate identification of therapeutically relevant targets is therefore essential to improve success rates and reduce resource waste. Traditional target identification heavily relies on laborious modeling experiments. These approaches often struggle to capture the full complexity of human diseases and their underlying molecular networks.

In recent years, the integration of multi-omics technologies with artificial intelligence (AI) has emerged as a transformative paradigm to overcome these limitations. Multi-omics, including genomics, transcriptomics, proteomics, and metabolomics, provides comprehensive molecular readouts of biological systems, enabling the identification of novel biomarkers, disease subtypes, and therapeutic targets (Chakraborty et al., 2018; Hurst and Knowles, 2018; Marshall et al., 2021). However, the volume, heterogeneity, and complexity of these data present analytical challenges that exceed the capabilities of conventional biostatistics (Bzdok et al., 2018). AI, particularly machine learning (ML), has emerged as a critical enabler for extracting biological insights from multi-dimensional omics datasets (Reel et al., 2021). ML models can integrate disparate data types, unravel complex molecular interactions, and prioritize high-value therapeutic targets with increasing accuracy (Chen et al., 2024a; Rodriguez et al., 2021; Schulte-Sasse et al., 2021; Tasaki et al., 2022). This integrated approach not only accelerates target identification and validation but also enhances lead optimization and clinical evaluation. The power of this synergy is exemplified by initiatives such as Insilico Medicine's AI-driven discovery of a novel target and corresponding drug candidate, which took only 18 months from target discovery to preclinical candidate nomination (Ren et al., 2024). Compared to the traditional approaches, the combination of multi-omics and AI accelerates the elucidation of disease mechanisms, expands the universe of druggable targets, and propels the advancement of personalized medicine (Dou et al., 2023; Zhang et al., 2023b).

Here, we provide an overview of the strategies and applications of integrating multi-omics with AI in drug development. We discuss key omics technologies, computational approaches, and data integration frameworks, with a focus on their use across the drug development pipeline, including target identification, validation, lead optimization, and clinical evaluation. We also explore current challenges and future directions in this rapidly evolving field.