22 Multi-omics
Multi-omics approaches integrate data from multiple sources. For example, we can integrate taxonomic abundance profiles with metabolomic or other biomolecular profiling data to observe associations, make predictions, or aim at causal inferences. Integrating evidence across multiple sources can lead to enhanced predictions, more holistic understanding, or facilitate the discovery of novel biomarkers.
In the following chapters, we explore common approaches for multi-assay data integration. These methods can be broadly categorized into association-based techniques, ordination models that leverage latent shared factors, and machine learning models for classification and regression. For more details, see the following chapters:
Back to top