Computational Methods for Pathway and Network Analysis with MicroRNAs

Pathway analysis is a key computational framework for interpreting high-throughput molecular data and contextualizing biological alterations within known cellular processes. The integration of microRNAs (miRNAs), key post-transcriptional regulators of gene expression, has substantially expanded the ability of pathway-based approaches to capture complex regulatory mechanisms underlying human diseases. Although methodologies such as Over-Representation Analysis (ORA), Functional Class Scoring (FCS), and Pathway Topology (PT)-based methods established the field's foundation, recent advances have profoundly reshaped both its methodological landscape and its biological scope. This chapter reviews the emerging computational paradigms driving the next generation of miRNA-integrated pathway analysis. In particular, it highlights three major developments: the rise of multi-omics integration, which enables the joint analysis of heterogeneous molecular layers such as transcriptomics, methylomics, and proteomics; the growing adoption of deep learning approaches, especially Graph Neural Networks (GNNs), for modeling complex and non-linear dependencies in biological networks; and the advent of single-cell sequencing, which provides unprecedented resolution for investigating cellular heterogeneity, dynamic regulatory programs, and causal relationships along pseudotemporal trajectories. The chapter also discusses recent advances in subpathway and network module analysis, which shift attention from individual pathways to coordinated network behavior and higher-order regulatory organization. Overall, these developments reflect a broader transition toward more integrative, dynamic, and mechanistically informative models of miRNA-mediated regulation, expanding our capacity to characterize complex biological systems.