The limma package is an R/Bioconductor software package designed for analyzing gene expression data from experiments. It provides tools for handling complex experimental designs and borrowing information to overcome small sample size issues. Over the past decade, limma has been widely used for differential expression analyses of microarray and high-throughput PCR data. The package now supports RNA-seq data, allowing similar analysis pipelines for both RNA-seq and microarray data. It also enables analysis of gene expression profiles in terms of co-regulated gene sets or higher-order expression signatures, enhancing biological interpretation. Limma integrates statistical principles to effectively analyze large-scale expression studies. It uses linear models to analyze entire experiments as a whole, sharing information between samples and modeling correlations. This approach allows for flexible hypothesis testing, including interaction effects and complex comparisons. Shared global parameters link gene-wise models, enabling information sharing between genes. Empirical Bayes methods are used to borrow information between genes, improving statistical power and accuracy. Limma allows for quantitative weights to account for varying data quality, and it supports RNA-seq data by converting counts to log-scale and estimating mean-variance relationships. It also handles unequal variability in expression data through mean-variance trends and precision weights. The package provides tools for preprocessing RNA-seq and microarray data, including background correction, normalization, and quality assessment. Limma includes functions for graphical exploration of data quality, such as diagnostic plots and mean-difference plots. It supports differential expression analysis, identifying genes that are differentially expressed between conditions. The package provides methods for testing differential expression, including empirical Bayes moderated t-statistics and P-values. It also supports differential splicing analysis using exon-level expression data. Higher-level analyses involve gene set testing, assessing the overall significance of co-regulated genes. Limma provides functions for gene set testing, including methods for adjusting for biases and testing for biological correlation between gene expression changes. The package also supports multiple testing adjustments, controlling for family-wise type I error rates or false discovery rates. Overall, limma offers a flexible and statistically rigorous framework for analyzing gene expression data across various platforms and experimental designs.The limma package is an R/Bioconductor software package designed for analyzing gene expression data from experiments. It provides tools for handling complex experimental designs and borrowing information to overcome small sample size issues. Over the past decade, limma has been widely used for differential expression analyses of microarray and high-throughput PCR data. The package now supports RNA-seq data, allowing similar analysis pipelines for both RNA-seq and microarray data. It also enables analysis of gene expression profiles in terms of co-regulated gene sets or higher-order expression signatures, enhancing biological interpretation. Limma integrates statistical principles to effectively analyze large-scale expression studies. It uses linear models to analyze entire experiments as a whole, sharing information between samples and modeling correlations. This approach allows for flexible hypothesis testing, including interaction effects and complex comparisons. Shared global parameters link gene-wise models, enabling information sharing between genes. Empirical Bayes methods are used to borrow information between genes, improving statistical power and accuracy. Limma allows for quantitative weights to account for varying data quality, and it supports RNA-seq data by converting counts to log-scale and estimating mean-variance relationships. It also handles unequal variability in expression data through mean-variance trends and precision weights. The package provides tools for preprocessing RNA-seq and microarray data, including background correction, normalization, and quality assessment. Limma includes functions for graphical exploration of data quality, such as diagnostic plots and mean-difference plots. It supports differential expression analysis, identifying genes that are differentially expressed between conditions. The package provides methods for testing differential expression, including empirical Bayes moderated t-statistics and P-values. It also supports differential splicing analysis using exon-level expression data. Higher-level analyses involve gene set testing, assessing the overall significance of co-regulated genes. Limma provides functions for gene set testing, including methods for adjusting for biases and testing for biological correlation between gene expression changes. The package also supports multiple testing adjustments, controlling for family-wise type I error rates or false discovery rates. Overall, limma offers a flexible and statistically rigorous framework for analyzing gene expression data across various platforms and experimental designs.