MetaNet: a scalable and integrated tool for reproducible omics network analysis

Network analysis is a powerful strategy for uncovering complex relationships in high-throughput omics datasets. However, current tools often lack scalability, flexibility, and native support for multi-omics integration, posing significant barriers for exploring complex biological networks. To address these limitations, we developed MetaNet, a high-performance R package designed to construct, visualize, and analyze biological networks from multi-omics datasets. MetaNet supports highly efficient correlation-based network construction, scalable to datasets with over 10,000 features, and includes extensive layout algorithms and visualization options compatible with both static and interactive platforms. It also provides a comprehensive suite of topological and stability metrics for in-depth network characterization. Benchmarking results show that MetaNet outperforms existing R packages by up to 100-fold in computation time and reduces memory usage by up to 50-fold. We demonstrate its utility through two case studies: (1) a longitudinal analysis of microbial co-occurrence networks showing the dynamics of the airborne microbiome, and (2) an integrative exposome-transcriptome network of more than 40,000 features, uncovering distinct regulatory impacts of biological and chemical exposures. MetaNet bridges the gap between network theory and omics application by offering a robust, reproducible, and biologically informed framework for large-scale, interpretable, and integrative network analyses across diverse omics platforms, advancing systems-level understanding in modern life sciences. MetaNet is available on the Comprehensive R Archive Network (https://cran.r-project.org/web/packages/MetaNet).