Simultaneous estimation of detection sensitivity and absolute copy number from digital PCR serial dilution

Xutao Deng; Brian S Custer; Michael P Busch; Sonia Bakkour; Tzong-Hae Lee

doi:10.1016/j.compbiolchem.2017.01.015

Simultaneous estimation of detection sensitivity and absolute copy number from digital PCR serial dilution

Comput Biol Chem. 2017 Jun:68:1-5. doi: 10.1016/j.compbiolchem.2017.01.015. Epub 2017 Feb 1.

Authors

Xutao Deng¹, Brian S Custer², Michael P Busch³, Sonia Bakkour⁴, Tzong-Hae Lee⁵

Affiliations

¹ Blood Systems Research Institute, San Francisco, CA 94118, USA; Department of Laboratory Medicine, University of California at San Francisco, San Francisco, CA 94107, USA. Electronic address: xdeng@bloodsystems.org.
² Blood Systems Research Institute, San Francisco, CA 94118, USA; Department of Laboratory Medicine, University of California at San Francisco, San Francisco, CA 94107, USA. Electronic address: bcuster@bloodsystems.org.
³ Blood Systems Research Institute, San Francisco, CA 94118, USA; Department of Laboratory Medicine, University of California at San Francisco, San Francisco, CA 94107, USA. Electronic address: mbusch@bloodsystems.org.
⁴ Blood Systems Research Institute, San Francisco, CA 94118, USA; Department of Laboratory Medicine, University of California at San Francisco, San Francisco, CA 94107, USA. Electronic address: sbakkour@bloodsystems.org.
⁵ Blood Systems Research Institute, San Francisco, CA 94118, USA; Department of Laboratory Medicine, University of California at San Francisco, San Francisco, CA 94107, USA. Electronic address: tlee@bloodsystems.org.

PMID: 28213308
DOI: 10.1016/j.compbiolchem.2017.01.015

Abstract

Digital polymerase chain reaction (dPCR) is a refinement of the conventional PCR approach to nucleic acid detection and absolute quantification. Digital PCR works by partitioning a sample of DNA or cDNA into many individual, parallel PCR reactions. Current quantification methods rely on the assumption that the PCR reactions are always able to detect single target molecules. When the assumption does not hold, the copy numbers will be severely underestimated. We developed a novel dPCR quantification method which determines whether the single copy assumption is violated or not by simultaneously estimating the assay sensitivity and the copy numbers using serial dilution data sets. The implemented method is available as an R package "digitalPCR".

Keywords: Copy number estimation; Digital PCR; Maximum likelihood; Poisson distribution; Serial dilution; Virus.

MeSH terms

Algorithms
DNA / genetics*
Polymerase Chain Reaction*

Substances

DNA