Model Selection for Correlation and Association Analysis
Yinglin Xia, Sun Jun · 2024
To effectively perform correlation and association analysis of multi-omics data (e.g., microbiome and metabolomics data), we need to reduce the dimensions of the data and find the significant microbial taxa and metabolites prior to applying correlation analysis. In Chapter 7 , we illustrate how to use the Kruskal–Wallis rank-sum test to identify significant microbial taxa and metabolites. First, we introduce the Kruskal–Wallis rank-sum test ( Section 7.1 ). Then, we perform model selection using the Kruskal–Wallis rank-sum test ( Section 7.2 ). Finally, we summarize this chapter in Section 7.3 .