Network analysis in systems nutrition

Network analysis can be useful to put the results of genetics analyses into biological context. This chapter reviews approaches to network analysis and their usefulness in integrating different data types in the study of the impact of nutritional interventions on biological systems. It details some of the most common networks, such as metabolic networks, protein-protein interaction networks, gene co-expression networks, and regulatory networks. Metabolic networks can be found in the Kyoto Encyclopedia of Genes and Genomes (KEGG), HumanCyc, Edinburgh Human Metabolic Network (EHMN), and Human Metabolic Reconstruction (Recon2) databases. Metabolomic data were associated with each single-nucleotide polymorphism (SNP) in the genetic dataset using genome-wide association study (GWAS), then significant SNP results were used as input to the VEGAS algorithm to determine gene-level R-values from SNP-level data. The chapter describes different studies in which network analysis was useful in analyzing results from GWAS, expression quantitative trait loci (eQTL), transcriptomics data and multi-omics studies.