Title:netgwas: An R Package for Network-Based Genome-Wide Association Studies

Abstract:Although Genome-Wide Association Studies provide a large number of single nucleotide polymorphisms (SNPs), the data are often analyzed one SNP at a time. The low power of single SNPs analysis, combined with the high significance threshold for performing multiple testing correction, substantially reduces the power of GWAS. Graphical models are powerful tools for modeling and making statistical inferences regarding complex associations among variables in multivariate data. They are capable of adjusting the effect of all other variables while measuring the pairwise associations. In this paper we introduce the R package netgwas, which is designed based on graphical models to accomplish three important and interrelated goals in genetics: constructing linkage map, reconstructing linkage disequilibrium networks from multi-loci genotype data, and detecting high-dimensional genotype-phenotype networks. The netgwas package can deal with species of any ploidy level. It implements recent improvements in both linkage map construction, and reconstructing conditional independence network for non-Gaussian data, discrete data, and mixed discrete-and-continuous data. Such datasets routinely occur in genetics and genomics such as genotype data, genotype-phenotype data, and multi-loci multi-trait multi-environment data. We demonstrate the value of our package functionality by applying it to five multivariate example datasets taken from the literature. We show, in particular, that our package allows a more realistic analysis of data, as accounts for simultaneous interactions among variables. This rules out spurious associations between variables that can arise from classical multiple testing approach. The package uses a parallelization strategy on multi-core processors to speed-up computations for large datasets. The netgwas package is freely available at this https URL.