This package is a gene set analysis function for one-sided test (OLS), two-sided test (multivariate analysis of variance). If the experimental conditions are equal to 2, the p-value for Hotelling's t^2 test is calculated. If the experimental conditions are great than 2, the p-value for Wilks' Lambda is determined and post-hoc test is reported too. Three multiple comparison procedures, Dunnett, Tukey, and sequential pairwise comparison, are implemented. The program computes the p-values and FDR (false discovery rate) q-values for all gene sets. The p-values for individual genes in a significant gene set are also listed. MAVTgsa generates two visualization output: a p-value plot of gene sets (GSA plot) and a GST-plot of the empirical distribution function of the ranked test statistics of a given gene set. A Random Forests-based procedure is to identify gene sets that can accurately predict samples from different experimental conditions or are associated with the continuous phenotypes.