dPCR assay development | Digital PCR troubleshooting

Digital PCR protocol — overview

Developing a digital PCR (dPCR) assay depends on many factors including the template, instrument used, applications and goals of the experiment. Most digital PCR protocols include a variation of these steps:

Sample preparation
–
dPCR assay

Sample preparation is a key step in achieving high PCR efficiency. There are several important parameters to consider when preparing the starting template for digital PCR.

These parameters include:

Sample purity
Sample integrity in terms of structure
Sample length and sequence
Sample input amount
Use of replicates
Use of controls

How to optimize these factors to achieve a better dPCR run is discussed below.

Pipetting as the first step of sample preparation in a dPCR protocol

Sample purity

Because PCR depends on multiple rounds of enzymatic reactions, it is more sensitive to impurities such as proteins, phenol/chloroform, salts and EDTA than single-step enzyme-catalyzed reactions. The purity of nucleic acid templates is important for dPCR because contaminants can interfere with fluorescence detection.

Alcohols (ethanol, isopropanol) and salts impair primer and probe annealing properties and reduce the efficiency of amplification, e.g., reduced fluorescence of positives and impeded discrimination of positive and negative partitions
Humic acids quench the fluorescence of dsDNA-binding dyes, e.g., EvaGreen
Nucleases degrade RNA and DNA
Urea and phenol denature the Taq polymerase
Acidic polysaccharides mimic nucleic acids and form dead-end complexes with the Taq polymerase

Despite being less prone to inhibitory effects than qPCR, dPCR works optimally using templates with high template purity. Various kits exist to achieve high nucleic acid purity and PCR efficiency depending on the template, such as genomic DNA, plasmid DNA, total RNA, etc.

Sample integrity: Length and sequence

Amplicon length and sequence have as much influence as template purity in determining the success of a dPCR experiment. Strongly degraded template RNA and DNA tend to show a discrepancy between OD-quantified DNA amount and the number of copies amplified and detected by dPCR. A larger-than-expected DNA amount might be necessary to achieve desired sensitivity (e.g., in mutation detection). It is advisable to keep amplicons as short as possible, particularly when using strongly degraded samples (FFPE DNA, cfDNA).

Similarly, residual crosslinks may prevent strand separation and amplification and abasic sites can induce unspecific amplification. Dedicated kits and protocols are available for the recovery of high-quality gDNA from FFPE samples.

Sample integrity: Structure

Random template partitioning is crucial for accurate quantification using a dPCR protocol. A uniform distribution of PCR signal is observed for most template DNA, including formalin-fixed, paraffin-embedded (FFPE) DNA, circulating cell-free DNA (cfDNA), complementary DNA (cDNA), and gBlocks.

To achieve uniform distribution of high-molecular-weight templates with complex structures, it is recommended to use strategies, such as restriction digestion.

Sample input amount

The sample input amount depends on the dPCR technology used. In QIAcuity nanoplate digital PCR, up to 217,000 copies per reaction can be used in 26k nanoplates and up to 170,000 copies in 8.5k nanoplates.

How to calculate the copy number

If the haploid genome size of an organism is known, the correlation between the mass input of genomic DNA (gDNA) and the resulting copy number (for a single-copy gene) can be calculated with the following formula:

Size of the genome (bp) x average weight of a single base pair (1.096 x 10^–21 g/bp)

For example, for the human genome with a genome size of approximately 3.3 x 10⁹ bp, the calculation is as follows:

3.3 x 10⁹ bp x 1.096 x 10⁻²¹ g/bp = 3.3 x 10⁻¹²g = 3.3 pg

The table below shows the copy number from 10 ng of gDNA from several model organisms

Organism	Genome size	Gene copies (1 copy/haploid genome) in 10 ng gDNA
Homo sapiens	3.3x10⁹	3000
Zebrafish	1.7x10⁹	5400
Saccharomyces cerevisiae	1.2x10⁷	760,500
Escherichia coli	4.6x10⁶	2,000,000
Standard plasmid DNA	3.5x10³	2,600,000,000

A note to remember

In dPCR, the average number of copies/partitions should not exceed 5. Ideally, it should be in the range of 0.5 to 3.

Replicates

It is recommended to analyze samples in duplicate or triplicate to prevent bias in quantification due to pipetting errors. Adding the data from duplicates together increases the number of measured events. This helps increase the precision of the digital PCR assay.

Controls

Negative controls – needed to monitor false-positive reactions, which could arise due to contamination or problems with the primers and probes. Negative controls are also used to determine the limit of detection (LOD).
Positive controls – used to test if the template amplification occurs under the set reaction conditions
Non-template controls (NTCs) – control contamination in all reagents

Digital PCR on the QIAcuity system: An introduction

The webinar covers various fundamental aspects of digital PCR on QIAcuity instruments, emphasizing choosing templates and their concentrations.

Watch now

Detection chemistry

DNA-binding dyes

Fluorescent intercalating DNA dyes, such as EvaGreen, bind all double-stranded DNA molecules and emit a fluorescent signal of a defined wavelength upon binding. Signal intensity increases with increasing cycle number due to the accumulation of PCR product. Using DNA-binding dyes such as EvaGreen enables the analysis of many different targets but eliminates the need to synthesize target-specific labeled probes. However, nonspecific PCR products and primer dimers could contribute to the fluorescent signal and even appear as a separate cluster of fluorescent signals during digital PCR data analysis. High PCR specificity is essential if using EvaGreen dye.

Fluorescence detection using DNA intercalating dyes in digital PCR

Fluorescence detection using primers/probes in digital PCR

Hydrolysis probes

Hydrolysis probes, called TaqMan probes, are sequence-specific oligonucleotides with a bound fluorophore and quencher moiety. A fluorophore binds to the 5' end of the probe, and a quencher binds to its 3' end or internally within the sequence of interest. During the extension phase of PCR, the probe is cleaved by the 5′→3′ exonuclease activity of Taq DNA polymerase, separating the fluorophore and the quencher moiety. As a result, a fluorescence signal proportional to the amount of accumulated PCR product can be detected.

As a dPCR troubleshooting tip, you should avoid certain combinations of reporter and quencher. For example, if the emission of the quencher overlaps with the emission of the fluorescent dye, additional background signals are created in the corresponding fluorescence channel. This background noise adversely affects cluster separation and peak resolution.

dPCR primer design

Effective primer and probe design are essential for the success of digital PCR assays. Primer and probe design in a dPCR protocol follows the same rules as for qPCR assays. The areas of focus during primer and probe design include:

Target matching
Base composition
Amplicon length
Melting temperature
Absence of secondary structures
Absence of self- and inter-complementarity (self-annealing)
Absence of cross-reactivity

Optimizing primer design and probe design for digital PCR assays

One difference between primer and probe design for a digital PCR protocol is that primer and probe concentrations in dPCR tend to be higher than in qPCR. Higher primer and probe concentrations increase the fluorescence intensity/amplitude, allowing for better separation of background noise from specific signals. Ultimately, more accurate target quantification can be achieved. Evidence suggests that optimal results are obtained at the final concentration of primer set between 0.5 µM – 0.9 µM and for probes at 0.25 µM per reaction.

Storage of primers and probes

Apart from careful assay design and the use of appropriate primer and probe concentrations, the correct storage of primers and probes is also critical to the success of dPCR.

Lyophilized primers and probes should be dissolved in a small volume of low-salt buffers, such as 100 µM TE buffer (10 mM Tris·Cl, 1 mM EDTA, pH 8.0) to give a concentration stock solution. As an exception, probes labeled with Cy5 and Cy5.5 fluorescent dyes should be stored in Buffer TE, pH 7.0 because they tend to degrade at higher pH.

To store the primers, small aliquots in nuclease-free TE buffer can be stored at -20°C for at least 1 year. Fluorescently labeled probes are stable under these same conditions for 6 to 9 months. Repeated freeze-thaw cycles should be avoided to reduce the risk of degradation.

For primer-probe sets used in dPCR multiplex assays, 20x primer-probe mixes could be used, with 2 primers and 1 probe for a particular target at the suggested concentrations.

To reconstitute primers and probes, a tube containing a lyophilized primer or probe should be centrifuged gently to collect all material at the bottom of the tube. The required volume of sterile, nuclease-free TE buffer should be added, mixed, and left for 20 minutes to allow the primer or probe to completely dissolve. The primer or probe solution should be mixed again and the concentration determined by spectrophotometry. A dPCR troubleshooting tip is to avoid dissolving the primers and probes in water. Some of these primers and probes have lower solubility and stability in water than in TE buffer.

Performing a dPCR run

A representation of the nanoplate dPCR workflow

There are three main reaction steps in a standard nanoplate dPCR workflow: Sample loading, amplification, and data analysis.

When optimizing dPCR reaction steps, qPCR conditions that generate quality qPCR results can often be directly applied to dPCR. However, checking the manufacturer’s recommendations for specific requirements is always a good idea.

Tips for sample loading

The first of the three dPCR reaction steps is to prepare and load the sample. During the preparation of the PCR mix and loading of the nanoplate, the following recommendations can be considered:

Decontaminate your workspace and labware to reduce the risk of foreign DNA contamination
To ensure maximal PCR efficiency, test the sample in different dilutions prior to the main digital PCR assay
Fully thaw all components before preparing reaction mixtures and mix components thoroughly to obtain homogeneous solutions and centrifuge briefly to avoid spillovers.
Combine master mix with the sample, primers, RNAase-free water and restriction enzymes if using any
Use sterile pipette tips
Pipette the reaction mixture into the nanoplate. Make sure no bubbles are introduced into the wells of the dPCR nanoplate during sample transfer and downstream transportation of the plate
To prevent evaporation and contamination, use a roller to carefully seal the nanoplate with foil
Carry nanoplate over to the dPCR instrument by only holding the side edges of the plate and avoiding shaking or turning to ensure the dPCR reaction mix stays at the bottom of the input wells
Set up software and start the dPCR run

Tips for amplification

The second of three dPCR reaction steps is to run the dPCR protocol through amplification. During this step, the reaction mix of each well is segregated into thousands of individual reactions. The PCR reaction is performed in a thermocycler. If template material is present within a partition, a positive fluorescence signal is detected during imaging. The images are processed by dedicated software. The current plate status can be monitored, typically on the dPCR instrument itself or using the connected software suite.

Tips for achieving optimal amplification conditions and PCR efficiency include:

Annealing temperature – can influence the dPCR assay specificity, typically set between 55°C and 65°C, the optimal annealing temperature is obtained when the largest separation between positive and negative partitions is achieved; increasing the annealing temperature might be beneficial in improving the separation between positive signals and background noise
Amplification cycles – it is recommended to run 40 amplification cycles in order to achieve sufficient separation between positive signals and background noise. In some cases, the number of thermal cycles may have to be increased to achieve optimum performance

Digital PCR data analysis

In the last of the three dPCR reaction steps, digital PCR data analysis is performed with absolute quantification based on Poisson statistics.

The QIAcuity Software Suite can perform digital PCR data analysis according to application aims: absolute quantification, mutation detection, genome editing, copy number variation or gene expression. In absolute quantification analysis (first-level analysis), the software generates a concentration diagram and a view of positive and negative partitions for selected wells. A heatmap view shows the target channel alongside the reference channel. Histograms and scatterplots can be used to change threshold settings and recalculate results. In the QIAcuity Software Suite, you can create reports about the analysis results of your plate. Absolute quantification is the prerequisite for all subsequent calculations and second-level analyses (e.g., mutation detection analysis, gene expression analysis, copy number variation analysis, etc.).

Tips for digital PCR data analysis:

Set a threshold for the classification of partitions as positive or negative just above the cluster of negative partitions
Use the NTC sample, with only negative partitions, to help in setting the threshold, but also perform an inspection of all wells
You can set the fluorescence amplitude of individual wells slightly higher or lower than NTCs to avoid misclassification of some partitions
Although no consensus value exists, a reaction is normally considered positive when the number of positive partitions exceeds 2
To address well-to-well and batch-to-batch variability, use volume precision factor (VPF) to define the exact volume of each well and apply the factor for concentration calculations by your software

Examples of digital PCR data

QIAcuity Digital PCR Applications Guide

Find in-depth information about setting up experiments and performing data analysis for dPCR applications.

Get guide

Digital MIQE (dMIQE) guidelines

Minimum Information for Publication of Digital PCR Experiments (dMIQE) guidelines, initially created to ensure the reproducibility and comparability of qPCR results, have also been adapted to the digital PCR protocol. The set of guidelines is used in the production and publication of dPCR data with the aim of harmonizing the dPCR approach and providing meaningful results, which can be readily interpreted across different laboratories.

The dMIQE guidelines provide a checklist of items for publishing results from dPCR experiments so that methods and results can be reproduced, followed and understood. The dMIQE guidelines are used to facilitate replication of experiments, and offer critical information for analysis and review of the technical quality of the work regardless of the dPCR technology used.

How to transfer and optimize assays from qPCR to dPCR

Considerations for the design of real-time PCR assays also apply to a digital PCR protocol. An assay that generates satisfactory qPCR results will also be successful in dPCR. It is still advisable to initially use concentrations from a validated qPCR method but to run a primer/probe concentration gradient if end-point fluorescence values of the positive partitions need improvement. As outlined in the sections above, attention should also be paid to the compatibility of fluorophores and recommended thermal cycling conditions and primer/probe concentrations for dPCR.

Parameters for dPCR assay optimization

Condition	Variable effect on	Range
Annealing temperature	•Specificity •Separation of positive and negative partitions •Assay artifacts	•Theoretical or working T_a +/– 2.5ºC
Primer/probe concentration	•Separation of positive and negative partitions (increase positive signal or decrease negative signal)	•0.8 µm primer; 0.4 µm probe (starting point)
Thermal cycling steps	•Remove intermediate partition signal	•2 or 3 (inclusion of 72ºC elongation)
Number of thermal cycle repeats	•Remove intermediate partition signal	•30–60 repeats

However, when onboarding a new assay, an initial pilot test is always recommended to assess performance. It is also advisable to use well-characterized and representative controls in an appropriate background matrix. The assay should be optimized to ensure analytical accuracy if performance is sub-optimal. In such cases, reporting details of the optimization process is critical for reproducibility.

For fast optimization of sub-optimal assays, by running a temperature gradient during the annealing steps, the QIAcuity master mixes can also be run on any real-time PCR instrument.

Whitepaper: Precision and sensitivity in qPCR vs. dPCR

Explore data comparing the performance of qPCR and dPCR to make an informed decision on which method suits your application.

Digital PCR troubleshooting

dPCR experiments can be affected by similar issues as qPCR or traditional or conventional PCR. However, there are some specific challenges that could arise when performing a digital PCR assay. The table below serves as a dPCR troubleshooting guide for commonly encountered problems in a typical dPCR protocol, outlining possible causes and our expert recommendations.

Problem	Possible causes	Our recommendations
No negative partitions	If NTC wells have negative partitions: The sample’s target concentration is too high	Run a dilution series to determine the optimal amount of input DNA
No negative partitions	If NTC wells have only positive partitions: Probe hydrolysis due to poor long-term storage of probe stock solution or premature probe cleavage by the polymerase	Reorder probe and make new probe stock solution or identify intra-assay interactions, redesign dPCR assay components to reduce binding and cleavage
No positive partitions	Restriction enzyme might have cut within the target sequence	Test the dPCR assay against DNA digested with a different restriction enzyme and undigested DNA
	The target sequence is part of a secondary structure or contains loops, repeats, high G/C content	Re-design for another target; use a restriction enzyme to fragment input DNA (without cutting in the target sequence itself)
	The amplification conditions are not optimal	Perform an annealing/extension temperature gradient to determine the optimal temperature for the digital PCR protocol; adjust physical parameters (elongation time, ramp rate, etc.)
	Partial dPCR inhibition	Change method/kit for purification of nucleic acids; consider adding facilitator, such as BSA
	Primers or probes do not match the target region as anticipated	Ensure no errors were made during dPCR assay design or synthesis of primers/probes
Positive signal from NTC wells	Template/amplicon contamination in the reagents	Wipe down pipettes, tip boxes and bench tops with 5–10% bleach; prepare samples and perform dPCR in separate rooms; wear appropriate personal protective equipment; dUTP containing mixes together with heat-labile uracil N-glycsolyase (UNG) could reduce number of false positives arising from contaminating PCR products, but have no effect on contaminants originating from the template
Low number of positives	Poor DNA purification	Use only purified DNA; try alternative method/purification kit
	Improper sample or nanoplate loading	Do not overload or underload sample DNA; pipette carefully, ensure entire dPCR mix has been transferred to nanoplate; ensure sample remains at bottom of the input wells
	Wrong primer concentration	Use recommended primer and probe concentrations
	Evaporation from nanoplate during amplification	Seal nanoplate carefully (also around edges) using roller and foil
	Bubbles in input wells;	Pipette carefully; transport nanoplate by holding on the side edges only; do not drop or invert nanoplate
	disturbance or improper sample transfer into nanoplate (so not all sample is collected at bottom of the input wells)
No clear separation between positive and negative partitions	Suboptimal amplification conditions	Increase the number of cycles (but do not exceed 50 cycles); decrease ramp rate (ex. 2°C/s); increase elongation time to 2 minutes, denaturation time to 1 minute (especially important for longer amplicons); perform a serial dilution to determine optimal amount of input DNA
Inability to separate positive and negative partitions when working with EvaGreen	Excessive amount of primer or DNA starting material	Use primers at recommended concentrations; do not overload starting material
	Fragment length is too long	Homogenize fragment length by restriction digestion
Appearance of background “rain” (partitions that fail to belong to the positive or negative population)	Nonspecific binding	Do not overload primers; try increasing annealing temperature, reducing number of cycles, decreasing extension and annealing time; ensure reagents are free of impurities
	Poor target accessibility	Use restriction digestion (avoid cutting your target sequence) or sonicating; to address RNA secondary structures: try to change target location or perform reverse transcription at warmer temperatures
	Late PCR onset due to partial inhibition of some partitions	Reduce amount of sample loaded; dilute sample further; change method or kit for nucleic acid purification
Appearance of noise (clusters of negative points above threshold)	Solid or bubble contaminants	Take extra care when preparing nanoplate; examine results manuall to make sense of them if possible
	Optical issues with instrument	Ensure your dPCR system has superior optics technology with high optics stability
Appearance of extra unexpected cluster	A sequence variant of the target of the interest	If the unexpected cluster is not desired: set threshold above cluster to exclude it from quantification, increase annealing temperature to improve specificity, digest with restriction enzyme to cut nonspecific target, re-design primer or probe to reduce complementarity to off-target sequence; if cluster represents a potential functional homologue, consider lowering the annealing temperature so the two clusters merge into one cluster
No concentration measurement in some wells	Problem with assembly of reaction mixture or sample preparation	Ensure dPCR assay is properly set up and DNA input and purity is satisfactory
	Poor handling of the nanoplate during transport (shaking, dropping)	Exercise extra care during sample preparation, pipetting dPCR reaction mix into nanoplate, handle nanoplate only by side edges, and avoid inverting, shaking, dropping prior to loading into dPCR system; if possible, manually set a threshold on the software to calculate a certain concentration
	Improper plate sealing (evaporation from sides)	Carefully seal the plate with foil using a roller
Measured concentrations are too low	Poor dPCR assay design	Run a temperature gradient experiment to make sure the dPCR assay is running at optimal conditions; verify fluorophore is not conjugated to a G residue; add primers and probes at recommended concentrations
	Poor target accessibility	Use restriction digestion (avoid cutting your target sequence) or sonicating
	Presence of PCR inhibitors in samples	Change method/kit for purification of nucleic acids; add facilitator, such as BSA
Large error bars (inconsistent results)	Poor mixing of reaction mixtures	Thoroughly mix the reaction mixtures
Large error bars (inconsistent results)	Problems with temperature uniformity of thermal cycler	Raise the denaturation temperature from 94 to 96°C for the first five cycles

More support with dPCR optimization

References

Iowa Institute of Human Genetics – Droplet Digital PCR (accessed January 20, 2023)

Lindner L et al. Reliable and robust droplet digital PCR (ddPCR) and RT-ddPCR protocols for mouse studies. Methods. 2021; 191(4):95-106.

Pecoraro S et al. Overview and recommendations for the application of digital PCR. EUR 29673 EN, Publications Office of the European Union, Luxembourg, 2019

QIAGEN. QIAcuity User Manual Extension. June 2021.

Schrader C, Schielke A, Ellerbroek L, Johne R. PCR inhibitors – occurrence, properties and removal. Journal of Applied Microbiology. 2012; 113(5):1014-1026.

Sidstedt M, Rådström P, Hedman J. PCR inhibition in qPCR, dPCR and MPS – mechanisms and solutions. Analytical and Bioanalytical Chemistry. 2020; 412:2009-2023.

The dMIQE Group and Hugget JF. The Digital MIQE Guidelines Update: Minimum Information for Publication of Quantitative Digital PCR Experiments for 2020. 2020; Clin Chem 66(8):1012-1029.

dPCR assay development and troubleshooting