Identification of genes for normalization of RT-qPCR gene expression data: A review of published literature

Reverse-transcriptase quantitative Polymerase Chain Reaction (RT-qPCR) is a well-established technique to quantify gene expression levels and critically depends on reference genes for data normalization. We performed a review of biomedical literature to analyse the usage of RT-qPCR in relation to other techniques for transcriptional analyses and to describe practices for the identification of suitable reference genes for RT-qPCR. In the 81 analysed studies, 3 genes (GAPD, ACTB, B2M) were included in ≥70% of cases, but ranked among the most stable genes in ≤1/3 of cases. The most frequently used normalizing algorithm was geNorm (83%), followed by NormFinder (73%) and BestKeeper (32%). We also analysed transparency and good laboratory practices based on adherence to the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines, using selected validated evaluation criteria. Overall, key MIQE criteria were satisfied in ≥50% of analyzed studies, but only four criteria (details of employed kit/enzyme for reverse transcription, priming method, primers/probes and DNA polymerase) were satisfied in ≥90% of cases. Data on assay repeatability were reported only in 15% of studies. The presence of pseudogenes as a potential confounder of assay specificity was evaluated only in 13% of studies. Finally, as few as 6% of studies accounted for the presence of known mutations of singly nucleotide polymorphisms when designing assay primers/probes. Better adherence to the MIQE guidelines should be encouraged. Publicly available transcriptomic and genomic data sets could be employed to refine the identification of suitable normalizing genes and to assist assay design.