Linear Methods for Analysis and Quality Control of Relative Expression Ratios from Quantitative Real-Time Polymerase Chain Reaction Experiments

Relative expression quantitative real-time polymerase chain reaction (RT-qPCR) experiments are a common means of estimating transcript abundances across biological groups and experimental treatments. One of the most frequently used expression measures that results from such experiments is the relative expression ratio (R(E)), which describes expression in experimental samples (i.e., RNA isolated from organisms, tissues, and/or cells that were exposed to one or more experimental or nonbaseline condition) in terms of fold change relative to calibrator samples (i.e., RNA isolated from organisms, tissues, and/or cells that were exposed to a control or baseline condition). Over the past decade, several models of R(E) have been proposed, and it is now clear that endogenous reference gene stability and amplification efficiency must be assessed in order to ensure that estimates of R(E) are valid. In this review, we summarize key issues associated with estimating R(E) from cycle threshold data. In addition, we describe several methods based on linear modeling that enable researchers to estimate model parameters and conduct quality control procedures that assess whether model assumptions have been violated.