Finding Differentially Expressed Genes (original) (raw)

The paper presents methodologies for identifying differentially expressed genes using statistical models that account for overdispersion in count data, specifically through the application of Negative Binomial and Poisson-gamma mixture distributions. Key tools such as edgeR, DESeq, and baySeq are discussed, highlighting their techniques for accurate variance estimation and the borrowing of information across genes. These approaches facilitate improved differential expression analysis in RNA-Seq data, essential for understanding gene regulation under varying biological conditions.