Abstract

An increasing number of microbiome studies are simultaneously collecting host omics profiles to better understand microbe-host interplay. Although, there remain analytical challenges due to the dimensionality, heterogeneity, and sparsity of the data, as well as the complexity of interactions within and between modalities. Measuring global, or multivariate, associations between the microbiome and a host omics modality is often the first step in integrative analysis. This is due to its ability to aggregate effects across all features and its meaningful biological interpretation in terms of ecological diversity. Accordingly, we propose a distance-based mutual information test of global association for microbiome multiomics integration. Mutual information is an advantageous metric, as it can capture general forms of dependence that most correlations fail to. We develop a k-nearest neighbor estimation procedure and corresponding permutation test for independence, as well as an ensemble distance metric. Using simulations studies and a microbiome-metabolome study of multiple sclerosis, we demonstrate that our method can detect microbiome multiomics associations that existing methods may miss.