On determining the power of digital PCR experiments
The experimental design that will be carried out to evaluate a nucleic acid quantification hypothesis determines the cost and feasibility of digital polymerase chain reaction (digital PCR) studies. Experiment design involves the calculation of the number of technical measurement replicates and the determination of the characteristics of those replicates, and this in accordance with the capabilities of the available digital PCR platform. Available digital PCR power analyses suffer from one or more of the following limitations: narrow scope, unrealistic assumptions, no sufficient detail for replication, lack of source code and user-friendly software. Here, we discuss the nature of six parameters that affect the statistical power, i.e., desired effect size, total number of partitions, fraction of positive partitions, number of replicate measurements, between-replicate variance, and significance level. We also show to what extent these parameters affect power, and argue that careful design of experiments is needed to achieve the desired power. A web tool, dPowerCalcR, that allows interactive calculation of statistical power and optimization of the experimental design is available.
KeywordsDigital PCR Power Design Replicates Variance
Compliance with ethical standards
Conflict of interest
Biogazelle provided support in the form of salaries for JV, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. MV and OT have no conflict of interest to declare.
- 6.Whale AS, Bushell CA, Grant PR, Cowen S, Gutierrez-Aguirre I, O’Sullivan DM, Zel J, Milavec M, Foy CA, Nastouli E. Detection of rare drug resistance mutations by digital PCR in a human influenza A virus model system and clinical samples. J Clin Microbiol 2016;54(2):392–400.CrossRefPubMedPubMedCentralGoogle Scholar
- 7.El Khattabi LA, Rouillac-Le Sciellour C, Le Tessier D, Luscan A, Coustier A, Porcher R, Bhouri R, Nectoux J, Sérazin V, Quibel T. Could digital PCR be an alternative as a non-invasive prenatal test for trisomy 21: a proof of concept study. PLoS One 2016;11(5):e0155009.CrossRefPubMedPubMedCentralGoogle Scholar
- 18.2018. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing Vienna Austria. https://www.R-project.org/.
- 21.Huggett JF, Garson JA, Whale AS. Digital PCR and its potential application to microbiology. Molecular microbiology: diagnostic principles and practice. In: Persing DH, Tenover FC, Hayden RT, Ieven M, Miller MB, Nolte FS, Tang Y, and van Belkum A, editors. Washington, DC: ASM Press; 2016. p. 49–57.Google Scholar
- 24.Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Stat Soc B Met 1995;57(1):289–300.Google Scholar
- 25.Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat 1979;6(2):65–70.Google Scholar