Decoding Drug Mechanisms of Action through Multi-Omics and Network-Based Modelling

Understanding how drugs exert their effects at the molecular and systems level is a fundamental challenge in pharmacology and computational biology. The ability to decode mechanisms of action (MoA) has been greatly enhanced by high-throughput multi-omics technologies, which generate comprehensive datasets spanning transcriptomics, proteomics, metabolomics, and genomics. However, despite the wealth of available biological data, integrating these heterogeneous sources to derive coherent mechanistic insights remains complex. Traditional MoA prediction models often rely on single-modal or structure-based features, which fail to capture the dynamic and interconnected nature of cellular processes. Recent breakthroughs in network biology and machine learning have provided the tools to integrate multi-omics information with biological networks, revealing emergent properties and drug effects that are otherwise obscured in linear analyses. Yet, there is still a gap in scalable, interpretable frameworks that can combine these modalities to model system-wide drug responses with precision and biological fidelity.

This Research Topic aims to advance the understanding and prediction of drug mechanisms of action by uniting multi-omics integration with network-based computational modelling. It seeks to develop and benchmark algorithms that can effectively map interactions between molecular layers, decode downstream biological pathways affected by therapeutic compounds, and derive interpretable systems-level models.

We welcome articles addressing, but not limited to, the following themes:

o Multi-omics data integration frameworks for mechanism of action inference

o Network-based machine learning and graph neural networks for drug response modelling

o Cross-platform and multi-scale biological data harmonization

o Computational frameworks for mapping drug–target and pathway interactions

o Network perturbation analysis and causal inference in pharmacogenomics

o Explainable AI and model interpretability for MoA discovery

o Benchmarking datasets and evaluation strategies for MoA prediction research

o Applications in drug repurposing, toxicity prediction, and personalized therapy design

o Multi-omics-driven biomarker discovery for personalized medicine in diverse populations.

o Role of microbiome in various chronic diseases.