GSVA is a gene set variation analysis method for microarray and RNA-seq data. It estimates pathway activity variation across samples in an unsupervised manner. GSVA is compared with existing sample-wise enrichment methods and demonstrates robustness in detecting subtle pathway activity changes. It works with both microarray and RNA-seq data and is an open-source R package part of Bioconductor. GSVA provides increased power to detect pathway activity changes compared to other methods. It is a starting point for building pathway-centric biological models and is useful for RNA-seq data analysis. GSVA is non-parametric and unsupervised, and can be applied to RNA-seq data. It is flexible and can be used for differential pathway activity and survival analysis. GSVA is compared with other methods in simulated data and real data, showing higher statistical power and accuracy. It is used for analyzing leukemia data and ovarian cancer survival data. GSVA is also applied to RNA-seq data, showing similar performance to microarray data. GSVA is a versatile method for analyzing gene set variation in both microarray and RNA-seq data.GSVA is a gene set variation analysis method for microarray and RNA-seq data. It estimates pathway activity variation across samples in an unsupervised manner. GSVA is compared with existing sample-wise enrichment methods and demonstrates robustness in detecting subtle pathway activity changes. It works with both microarray and RNA-seq data and is an open-source R package part of Bioconductor. GSVA provides increased power to detect pathway activity changes compared to other methods. It is a starting point for building pathway-centric biological models and is useful for RNA-seq data analysis. GSVA is non-parametric and unsupervised, and can be applied to RNA-seq data. It is flexible and can be used for differential pathway activity and survival analysis. GSVA is compared with other methods in simulated data and real data, showing higher statistical power and accuracy. It is used for analyzing leukemia data and ovarian cancer survival data. GSVA is also applied to RNA-seq data, showing similar performance to microarray data. GSVA is a versatile method for analyzing gene set variation in both microarray and RNA-seq data.