Savitzky-Golay smoothing and differentiation for PCR quantification

In quantitative PCR (qPCR), replicates can minimize the impact of intra-assay variation; however inter-assay variations must be minimized in order to obtain a robust quantification method. The method proposed in this study uses Savitzky-Golay Smoothing and Differentiation (SGSD) to identify a derivative maximum-based cycle of quantification. It does not rely on curve modeling, as is the case with many existing techniques. PCR fluorescence data sets challenged for inter-assay variations (different thermocycler units, different reagents batches, different operators, different standard curves and different labs) were used for the evaluation. The algorithm was compared to a four-parameters logistic model method (4PLM), the Cy0 method and the threshold method. The SGSD method compared favorably to all methods in terms of inter-assay variation. SGSD was statistically different from 4PLM (p-value = 0.03), Cy0 (p-value = 0.05) and threshold method (p-value = 0.004) on relative error comparison basis. For intra-assay variations, SGSD outperformed the threshold method (p-value = 0.005) and equalled the 4PLM and Cy0 methods (p-value > 0.05) on relative error basis. Our results demonstrate that the SGSD method could potentially be an alternative to sigmoid modeling based methods (4PLM and Cy0) when PCR data are challenged for inter-assay variations.The accepted manuscript in pdf format is listed with the files at the bottom of this page. The presentation of the authors' names and (or) special characters in the title of the manuscript may differ slightly between what is listed on this page and what is listed in the pdf file of the accepted manuscript; that in the pdf file of the accepted manuscript is what was submitted by the author