Droplet Volume Variability and Impact on Digital PCR Copy Number Concentration Measurements

Digital polymerase chain reaction (dPCR) is increasingly being adopted by reference material producers and metrology institutes for value assignment, and for homogeneity and stability studies of nucleic acid reference materials. A reference method procedure should fulfill several requirements, and the uncertainty and biases should be completely understood. A bias in target concentration when inaccurate droplet volume is used in the droplet dPCR measurement equation has previously been documented. In this study, we characterize both intrawell and interwell droplet volume variability using optical microscopy and determine the impact of these two sources of variability on target concentration estimates. A small optical distortion across the image was measured which, without correction, biased droplet volume measurements. Longitudinal monitoring of interwell droplet volume over 39 weeks using several lots of Mastermix demonstrated a mean droplet volume of 0.786 nL and intermediate precision of 1.7%. The frequency distribution of intrawell droplet volumes varied. Some wells displayed a skewed distribution which resulted in a small bias in estimated target concentration for a simulated dPCR with target concentrations of between 62 and 8000 copies mu L-1. The size and direction of this bias was influenced by the distribution pattern of the droplet volumes within the well. The proportion of Mastermix in dPCR mix affected droplet volume. A pipetting error of 10% during mixing of the premix and Mastermix resulted in a 2.6% change in droplet volume and, consequently, a bias in concentration measurements highlighting the advantages of gravimetric preparation of dPCR mixes for high accuracy measurements.