geNorm

geNorm is a statistical algorithm and software tool used to identify and validate stable reference genes for normalizing RT-qPCR data. Introduced by Vandesompele and colleagues in 2002, it provides an objective approach to selecting internal control genes from a set of candidates.

The core of geNorm is the pairwise comparison of candidate reference genes across all samples. For each gene, it computes an expression stability measure, M, defined as the average pairwise variation of that gene with all other candidate reference genes. A lower M value indicates more stable expression. Genes are ranked by increasing M, and the least stable gene is removed in an iterative process until a preferred number of reference genes remain. The normalization factor is then calculated as the geometric mean of the relative quantities of the selected stable reference genes.

To determine how many reference genes are necessary, geNorm calculates pairwise variation values, V_n/n+1, comparing normalization factors based on n versus n+1 genes. A common guideline is that if V_n/n+1 falls below 0.15, adding another reference gene does not substantially improve normalization, though this threshold is not universal and dependent on the dataset.

Practical considerations include ensuring accurate amplification efficiencies, using non-co-regulated candidate genes, and employing a representative sample set. The method emphasizes using multiple reference genes rather than a single control to improve normalization reliability. geNorm has become widely used in gene expression studies as a standard approach for selecting stable reference genes.