Figure 1

The distribution of polymerase chain reaction (PCR) efficiencies, calculated as the differences between the Ct values of undiluted sample and the Ct values of a 10-fold diluted sample. (A) Theoretically, this difference should equal 3.32 or 2log(10). All miRNA assays with a difference in Ct value between 3.32% and 25% were included for further analysis. A blue dashed line indicates the boundaries of the interval; a blue solid line indicates the theoretical expected value of 3.32. To account for differences in preamplification, we compared the Ct values of a sample before and after preamplification. The median difference in Ct value was 8, and all miRNA assays with a difference of 8% ± 25% were included for further analysis. The scatterplot in (B) demonstrates an almost perfect linear relation for those selected miRNAs before and after preamplification. The blue line represents the regression line, for which the equation is given on top of the scatterplot. To evaluate assay reproducibility, we tested four samples in duplicate. The scatterplot in (C) demonstrates the result for one of these samples. The blue line represents the regression line, and the correlation coefficient resulting from the comparison of both profiles is given on top of the scatterplot. Further technical validation of our miRNA-expression data was performed for 12 samples by analyzing their miRNA-expression profile with the nCounter Analysis System and comparing this result with the qRT-PCR-based miRNA-expression profile. The scatterplot in (D) demonstrates the result for one of these samples. The blue line represents the regression line, and the correlation coefficient resulting from the comparison of both profiles is given on top of the scatterplot.