Comparison of the 2-CT method and the 2-ΔΔCT method for real-time qPCR data analysis

Abstract

[purpose]The [Formula] method proposed by Livak and Schmittgen in 2001 is used for real-time quantitative polymerase chain reaction (RT-qPCR) data analysis. This methods fundamental logic involves normalizing all data against the control group to achieve relative quantification of gene expression. During our practical calculations, we identified an inherent bias in this method: the practice of directly taking arithmetic means of a set of C (or {Delta}C) values during [Formula] calculations fails to fully account for the exponential nature of CT values (where CT values serve as exponents in power-of-2 operations), leading to deviations in computational results. In this study, we elucidate this systematic bias and propose an innovative approach to circumvent it. [Methods]We propose the [Formula] method, which bases calculations on [Formula] values and computes the fold-change of target genes relative to control group target genes. The computational logic of the [Formula] method strictly adheres to the exponential characteristics of CT values, avoiding the use of arithmetic averaging at CT and {Delta}CT levels, thereby more accurately reflecting gene expression levels in datasets. In this paper, we detail the computational process of the [Formula] method and compare results from both methodologies. [Results]Calculations based on Livak and Schmittgens published data show differences between the two methods, though the discrepancies are relatively small. In calculations from our recent cadmium exposure experiments, the [Formula] method indicates that 8-hour cadmium exposure increases irg-6 gene expression in C. elegans from 1.314 to 7.125-fold, while the [Formula] method shows an increase from 1 to 4.124-fold. The two methods exhibit nearly 70% discrepancy in irg-6 gene fold-change quantification, with statistical p-values of 0.0002 and 0.0015 respectively. In all calculations presented in the paper, the [Formula] method fails to produce an exact unit value (mean fold-change = 1) for control groups, whereas the [Formula] method consistently yields control group results of exactly 1, demonstrating its capability for precise normalization. [Conclusions]In summary, the [Formula] method demonstrates superior computational rigor, and we recommend its adoption in RT-qPCR data analysis. This improvement proves particularly valuable for experimental datasets with substantial CT value variability among different samples, establishing a more reliable computational paradigm for RT-qPCR analysis.