Droplet volume variability as a critical factor for accuracy of absolute quantification using droplet digital PCR
Anal Bioanal Chem
Droplet volume variability as a critical factor for accuracy of absolute quantification using droplet digital PCR
Alexandra Bogožalec Košir 0 1 2
Carla Divieto 0 1 2
Jernej Pavšič 0 1 2
Stefano Pavarelli 0 1 2
David Dobnik 0 1 2
Tanja Dreo 0 1 2
Roberto Bellotti 0 1 2
Maria Paola Sassi 0 1 2
Jana Žel 0 1 2
0 Instituto Nazionale di Ricerca Metrologica , Strada delle Cacce, 91-10135 Turin , Italy
1 Jožef Stefan International Postgraduate School , Jamova 39, 1000 Ljubljana , Slovenia
2 Department of Biotechnology and Systems Biology, National Institute of Biology , Večna pot 111, 1000 Ljubljana , Slovenia
Accurate and precise nucleic-acid quantification is crucial for clinical and diagnostic decisions, as overestimation or underestimation can lead to misguided treatment of a disease or incorrect labelling of the products. Digital PCR is one of the best tools for absolute nucleic-acid copy-number determ i n a t i o n . H o w e v e r, d i g i t a l P C R n e e d s t o be w e l l characterised in terms of accuracy and sources of uncertainty. With droplet digital PCR, discrepancies between the droplet volume assigned by the manufacturer and measured by independent laboratories have already been shown in previous studies. In the present study, we report on the results of an inter-laboratory comparison of different methods for droplet volume determination that is based on optical microscopy imaging and is traceable to the International System of Units. This comparison was conducted on the same DNA material, with the examination of the influence of parameters such as droplet generators, supermixes, operators, intercartridge and intra-cartridge variability, and droplet measuring protocol. The mean droplet volume was measured using a QX200™ AutoDG™ Droplet Digital™ PCR system and two QX100™ Droplet Digital™ PCR systems. The data show significant volume differences between these two systems, as well as significant differences in volume when different supermixes are used. We also show that both of these droplet generator systems produce droplets with significantly lower droplet volumes (13.1%, 15.9%, respectively) than stated by the manufacturer and previously measured by other laboratories. This indicates that to ensure precise quantification, the droplet volumes should be assessed for each system.
Droplet digital PCR; Droplet volume; DNA quantification; Optical microscopy imaging
-
Alexandra Bogožalec Košir and Carla Divieto contributed equally to this
work.
Introduction
Accurate and precise nucleic-acid quantification is
fundamental in many fields, from basic research to molecular
diagnostics, and in preclinical and clinical research and industrial
processes. Absolute DNA quantification allows the reduction of
measurement bias between laboratories, which is essential for
target DNA measurements in medicine, viral load analysis in
diagnostics, and microbial quantification in microbiology.
Currently, the most used method for DNA quantification is
real-time polymerase chain reaction, which depends on the
standard curve approach for quantification. However, the
choice of the standard can in some fields be arbitrary and is
often not uniform among laboratories [
1
]. Absolute
quantification of nucleic acids without the need of a standard curve
can be achieved with digital PCR (dPCR). Taking into
account the high potential of absolute quantification by dPCR,
this technique represents a good candidate for reference
methods for nucleic-acid copy-number determination [
1, 2
].
Although the concept of dPCR was defined in the 1990s [
3,
4
], dPCR has gained further popularity in more recent years
[
1, 5–14
], with 442 papers listed in the Scopus database for
2016, compared to 125 in 2012 (Fig. S1 in the Electronic
Supplementary Material, ESM). The main characteristic of
dPCR is the dispersal of the reaction volume into partitions,
either chambers in chamber/microfluidics-based digital PCR,
or droplets in droplet digital PCR (ddPCR). The underlying
assumptions of dPCR which need to be met include a random
distribution of DNA molecules into the partitions and
amplification of a single target in a partition [
15, 16
]. To achieve
this, the sample and reaction preparation need to be carefully
controlled, and the dPCR needs to be well characterised in
terms of accuracy and sources of uncertainty.
In dPCR, the copy number concentration in the sample is
calculated according to Eq. (1),
Tc ¼ −ln 1−
P
R
1
Vd
D
ð1Þ
where Tc is the mean target concentration (copies/μL), P is the
number of positive partitions, and R is the number of analysed
partitions. The copy number concentration depends also on
the partition volume (Vd) and the dilution factor of the original
solution in the PCR reaction (D). Consequently, the accuracy
of absolute quantification is strongly dependent on correct
determination of the partition volume, which is the droplet
volume for ddPCR, and on the number of partitions. The
confid (...truncated)