Drug Discovery from Gene Expression/Gene Regulation

Gene expression and gene regulation play critical roles in understanding disease mechanisms, identifying therapeutic targets, and guiding drug discovery. Advances in high-throughput sequencing technologies, such as next-generation sequencing (NGS) and single-cell RNA sequencing (scRNA-seq), have enabled comprehensive profiling of gene expression across diverse biological contexts. Computational methods, including machine learning (ML) and deep learning (DL), have further revolutionized transcriptomic and regulatory analyses by uncovering complex gene interactions and predicting disease-associated regulatory networks. This chapter explores the integration of AI-driven approaches in gene expression analysis, focusing on differential expression analysis, clustering techniques, and gene regulatory network (GRN) inference. We discuss the role of dimensionality reduction and unsupervised learning in identifying disease subtypes and patient stratification, as well as how single-cell transcriptomics enhances our understanding of cellular heterogeneity in drug response and resistance. Furthermore, we examine network-based methodologies, highlighting the application of AI and ML in reconstructing GRNs, prioritizing regulatory targets, and facilitating drug repositioning. The emerging field of AI-assisted drug repositioning is also covered, with an emphasis on leveraging transcriptomic signatures and network-based inference to repurpose existing drugs for novel therapeutic applications. By integrating multiomics data and employing explainable AI frameworks, researchers can refine precision medicine strategies and accelerate the drug discovery process. Finally, we address current challenges in AI-driven gene analysis, including data complexity, model interpretability, and the integration of multiomics datasets. Future directions include advances in large language models (LLMs), spatial transcriptomics, and quantum-enhanced AI for predictive modeling. The convergence of AI, high-throughput sequencing, and network-based approaches promises to reshape biomedical research and optimize drug discovery pipelines for improved clinical outcomes.