The quality control method employed in Kineret is a binary classification test based on comparing amplification kinetics of tested PCR reactions with reference PCR reactions.

Reactions that are found incompatible with the reference set are considered kinetics outliers. Note that the term reference applies here to a set of PCR reactions with assumed good performance rather than to reference gene.

Frequent reasons for kinetics outliers are for instance inhibition, non-specific amplification, missing template, missing reaction component, or an excess of reaction component. The amplification kinetics of an individual reaction is described by the maxima of the first and the second order derivative of the smoothed sigmoid amplification trajectory. Based on these two parameters, multivariate kinetics distance from the centre of the reference set is calculated for each reaction. Note that indication of outliers is a statistical process. The kinetics distance has approximately chi-square distribution. This means that there is a defined risk of false outliers. Therefore the critical kinetics distance for chosen false positive risk can be obtained from table 1.

Critical distance associated with fals positivity

To obtain high specificity, reactions should be compared to a reference set that is ideally composed of validated samples, for instance standard curve samples. Where no such reference set can be used, individual reactions can be compared to the total set of all reactions.

Note that the specificity and sensitivity of the test depends on the size and the quality of the reference set. Small (N<5) or poor quality reference set with inconsistent amplifications will produce outcomes with low sensitivity. In contrary, using reference set consisting of technical replicates only will produce KOD test with elevated false positivity. We recommend preparing reference from naturalistic samples with different concentrations of template covering the quantification range.

The multivariate KOD test has very high sensitivity to different sample matrix. The sample matrix of the reference reactions should therefore be similar to the sample matrix of the test reactions. Note that in order to achieve maximum performance of the KOD test, as many as possible reference PCR reactions should be used within a single test. Therefore if possible, do not split one gene over different runs. Alternatively, PCR reactions with the same gene performed in different runs can be imported together as if they were conducted in a single run. It is then recommended to test the inter-run homogeneity by declaring all wells of one run as reference and then running the KOD test. Only if the inter-run homogeneity is assured, the overall KOD test should be run.