Standard Curve
A dilution series of a sample with known concentration can be measured and used to create a standard curve that relates Cq values to concentration. The concentrations are defined in a classification column as shown below.
Open the Regression tab among the analysis tabs in the top of the main window, and press the Standard curve button to load the analysis into the Control panel.
Select the gene measurements you want to use for calibration in the Gene (Y) drop-down list, and the classification column that contains the concentration values in the Concentration (X) drop-down list. Options are available to find confidence intervals with confidence level of 90%, 95%, or 99%. The input is expected in logarithmic scale e.g. Cq values. If you have data on linear scale, such as copy numbers, apply a logarithm to your data before analysis by ticking the Apply log on (X) check box. Press the Run button to get the results.
A standard curve (blue line) is obtained by fitting a regression line to the samples with known concentrations. A confidence interval (red lines on either side of the standard curve) is calculated based on these data and on the confidence level that was selected in the Control panel. The samples that the regression line is based on are indicated with dots in the resulting plot.
Under normal conditions the measured sample point will deviate from the regression line. The amplitude of deviation can be used as a means to estimate the quality of the fit. The residual plot shows the amplitude of deviation for each sample point as a function of the sample points concentrations.
A table holds the results and confidence intervals from the analysis. The orange columns contain the sample concentrations (X) and sample measurements (Y). Average and No. of samples are reported below these two columns. The column with grey background (Y-Ý) contains the residuals (deviations from the standard curve) from the regression line fit.
The calculated standard curve parameters are the Slope, the Intercept at zero concentration, and qPCR Efficiency which are given at the bottom of the table in yellow cells. They are presented with confidence interval of a given confidence level (Confidence) in pink cells. The statistical parameters that was used to calculate these confidence intervals are given at the bottom of the spreadsheet: Residual variation, standard error of the intercept (SE(intercept)), slope (SE(slope)), and efficiency (SE(Efficiency)), the selected confidence limit (Confidence), and the Critical t-value under these conditions.
There are two statistical test being performed and the result is presented in the table. The AS(Y-Ý) is the autoscaled Y-Ý and you can see the Mean and standard deviation (SD) of the Y-Ý below the autoscaled column. Grubbs' test statistic (Z(Grubbs test))indicates the largest deviation from the mean (AS(Y-Ý)) before a sample can be classified as an outlier. The outliers are indicated in pink in the AS(Y-Ý) column.
There is also a test for randomness in the data: Runs test. The null hypothesis is that the samples are random and independent of each other. Columns T+, T-, and Runs (colored teal) as well as cells m and s are used to calculate a p-value (P(RUNS test)). A low p-value indicates that the data is not random.
