Modelling the limit of detection in real-time quantitative PCR

The limit of detection (LOD) is a critical performance characteristic of an assay that requires careful evaluation during method validation. However, formal calculations for the LOD do not take into account atypical data sets that are generated from real-time PCR techniques, which can be non-normally distributed, truncated, and heteroscedastic. Experimental data sets for the quantification of genetically modified (GM) material were produced using real-time PCR, in order to model the LOD. A bootstrapping computer simulation calculated the probabilities of detecting PCR positive test results from these data sets, and computer modelling defined a function from the resulting probability plots. The LOD was modelled as a function of sample replication level and cycle threshold values. The broad applicability of this bootstrapping and data modelling approach should be of general interest to laboratories conducting trace-level detection.