Network integration is a knowledge-based analytical approach that map statistically identified omics features into curated biological interaction and pathway networks to provide functional and system-level interpretation. This enables users to interpret multi-omics results within biologically meaningful structures such as pathways, functional modules, and regulatory subnetworks.
OmicsAnalyst 2.0 provide two ways to perform network integration
-
Direct feature list upload:
Users can upload lists of omics features, including genes, proteins, metabolites, miRNAs,or microbial taxa, obtained from external experiments or other analysis. These features are mapped onto curated interaction and pathway networks to construct biological networks for functional annotation, module detection, and pathway enrichment. -
Projection from statistical analysis pipeline:
OmicsAnalyst 2.0 also supports direct projection of statistical results into the knowledge-based network layer, including statistically significant features, correlated cross-omics feature pairs, and top-loading features derived from multi-omics integration models.
Through network integration, users can discover pathways associated with specific phenotypes or experimental conditions, identify highly connected or central molecular entities that may represent key regulatory drivers, detect functional modules acting as coordinated biological units, or characterize cross-omics regulatory relationships. Together, these insights enable biologically coherent interpretation of multi-omics results and support hypothesis generation for downstream statistical and causal validation.