Optimal use of statistical methods to validate reference gene stability in longitudinal studies

Abstract

Multiple statistical approaches have been proposed to validate reference genes in qPCR assays. However, conflicting results from these statistical methods pose a major hurdle in the choice of the best reference genes. Recent studies have proposed the use of at least three different methods but there is no consensus on how to interpret conflicting results. Researchers resort to averaging the stability ranks assessed by different approaches or attributing a weighted rank to candidate genes. However, we report here that the suitability of these validation methods can be influenced by the experimental setting. Therefore, averaging the ranks can lead to suboptimal assessment of stable reference genes if the method used is not suitable for analysis. As the respective approaches of these statistical methods are different, a clear understanding of the fundamental assumptions and the parameters that influence the calculation of reference gene stability is necessary. In this study, the stability of 10 candidate reference genes ( Actb , Gapdh , Tbp , Sdha , Pgk1 , Ppia , Rpl13a , Hsp60 , Mrpl10 , Rps26 ) was assessed using four common statistical approaches (GeNorm, NormFinder, Coefficient of Variation or CV analysis and Pairwise ΔCt method) in a longitudinal experimental setting. We used the development of the cerebellum and the spinal cord of mice as a model to assess the suitability of these statistical methods for reference gene validation. GeNorm and the Pairwise ΔCt were found to be ill suited due to a fundamental assumption in their stability calculations. Highly correlated genes were given better stability ranks despite significant overall variation. NormFinder fares better but the presence of highly variable genes influences the ranking of all genes because of the algorithm’s construct. CV analysis estimates overall variation, but it fails to consider variation across groups. We thus highlight the assumptions and potential pitfalls of each method using our longitudinal data. Based on our results, we have devised a workflow combining NormFinder, CV analysis along with visual representation of mRNA fold changes and one-way ANOVA for validating reference genes in longitudinal studies. This workflow proves to be more robust than any of these methods used individually.

Methodological considerations on selection of stable reference genes for RT-qPCR in the neonatal rat brain in hypoxia and hypothermia

“…NormFinder calculates the stability score (S) based on the inter- and intra-group variation. However, it has been reported that including genes with high overall variation can affect the stability ranking of all genes with this method (22). This algorithm can potentially be improved after identifying and removing genes with high overall variation.. Actb , Sdha and Pgk1 were the most stable RGs, presented stability values lower than 0.3.…”

“…This makes the identification of the best RGs very unwieldy. Using the same statistical methods, new approaches have been proposed, such as the “Integrated approach” (22) that has shown to provide a more accurate estimate of RG stability. It is advisable to devise integrated approaches based on suitability for each experimental setting.…”

“…To address this issue, several approaches that make use of several statistical approaches at the same time, have been proposed, including i) a weighted rank (18–20), an approach that is compromised by the fact that it does not consider the strengths and drawbacks of each method for a given experimental setting; ii) the “Geometric mean rank” that uses the average of the stability ranks across different methods yielding an overall ranking (12,21); as well as iii) an integrated approach (22), including a first selection step making use of the CV analysis (eliminating genes with CV>50%), and subsequently ranking the remaining genes using GeNorm.…”

Selection of Reliable Reference Genes for Analysis of Gene Expression in Spinal Cord during Rat Postnatal Development and after Injury

“…The highest popularity acquired the "ready-to-use" user-friendly applets geNorm, NormFinder, and BestKeeper [19][20][21]. In spite of their drawbacks and occasional inconsistency in results, when compared side by side [15,27], these applets represent straightforward and valuable tools for routine RG testing. The source of the abovementioned inconsistency between the results of the applets lies in their unique algorithms assuming relative stability of the tested gene expressions.…”

“…To our knowledge, we are not aware of a serious survey and validation of stably expressed RGs in the rat spinal cord across postnatal development or after SCI. The issue was partially addressed by the study of Sundaram et al [15], who focused on the appropriate use of methods validating RGs' stability in the murine spinal cord. Several other papers deal with the validation of stable RGs in the spinal cord by neuropatic pain after spared nerve injury (SNI) and peripheral nerve injury (PNI) or after inflammatory injury [16][17][18].…”