Gaussian graphical models with applications to omics analyses

Gaussian graphical models (GGMs) provide a framework for modeling condi- tionaldependenciesinmultivariatedata.Inthistutorial,weprovideanoverview of GGM theory and a demonstration of various GGM tools in R. The math- ematical foundations of GGMs are introduced with the goal of enabling the researcher to draw practical conclusions by interpreting model results. Back- ground literature is presented, emphasizing methods recently developed for high-dimensionalapplicationssuchasgenomics,proteomics,ormetabolomics. Theapplicationofthesemethodsisillustratedusingapubliclyavailabledataset of gene expression profiles from 578 participants with ovarian cancer in The CancerGenomeAtlas.Stand-alonecodeforthedemonstrationisavailableasan RMarkdownfileathttps://github.com/katehoffshutta/ggmTutorial.