A well-designed and optimized ddPCR reaction yields highly reproducible and robust results. Before running a ddPCR experiment, it is important to understand the goal or expected outcome of the experiment, as different types of experiments require different sample preparation methods, different amounts of input target DNA, and different types of data analysis. Here, we discuss the fundamental experimental parameters to consider in designing a ddPCR assay. Many factors, such as amplicon selection and primer and probe design, will be familiar to researchers experienced in designing qPCR assays, while others are unique to the ddPCR system. This section explores ddPCR assay design, selecting target sequences, designing primers and probes, and sample preparation for a ddPCR assay.
PCR parameters play an important role in obtaining accurate results. A successful PCR reaction requires efficient and specific amplification of the product. Because the properties of both the primers and the target sequence can affect amplification efficiency, care must be taken when choosing a target sequence and designing primers.
Target amplification in a ddPCR system follows the same principles as in qPCR and regular PCR assays. Primer and probe hybridization kinetics, as well as polymerase processing, follow similar patterns whether amplification is in a 20 µl bulk reaction or in nanoliter droplets. This similarity may not be obvious at first glance, but at a molecular level, the relative numbers of reactants are still large; in a nanoliter-sized drop, there are approximately 540 million molecules of each primer (900 nM) and 150 million probe molecules (250 nM).
Digital PCR assays focus on the end-point analysis of each partition (droplet) to generate quantitative data. As such, amplification efficiency plays a smaller role in the outcome of the results than in qPCR; nevertheless, a proper assay should be designed and sometimes optimized to allow for clear and crisp differentiation of positive and negative droplets.
When choosing a region of the target for amplification, follow these guidelines:
- Plan to amplify a 75–200 bp product. Short PCR products are typically amplified with higher efficiency than longer ones, but a PCR product should be at least 75 bp long to allow its discrimination from any primer-dimers that may form
- Avoid regions that have secondary structure when possible. Use programs such as mfold to predict whether primers will anneal to form a secondary structure or a PCR product will form any significant secondary structure at the annealing temperature
- Avoid regions with long (>4) repeats of single bases
- Choose a region that has a GC content of 50–60%
When designing primers for a chosen target sequence, follow these guidelines:
- Design primers that have a GC content of 50–60%
- Strive for a Tm between 50 and 65°C. Calculate Tm values using the nearest-neighbor method, with values of 50 mM for salt concentration and 300 nM for oligonucleotide concentration
- Avoid secondary structure; adjust primer locations so they are outside the target sequence secondary structure, if required
- Avoid repeats of Gs or Cs longer than 3 bases
- Place Gs and Cs on the ends of the primers
- Check the forward and reverse primer sequences to ensure that there is no 3' complementarity (to avoid primer-dimer formation)
Verify the specificity of the primers for the target sequence using tools such as the Basic Local Alignment Search Tool (BLAST). Ensure there are no SNPs (single nucleotide polymorphisms) within the primer sequences.
A number of free online resources are available to facilitate primer design, including the Primer3 website (Whitehead Institute for Biomedical Research, MIT). Commercially available programs such as Beacon Designer Software (Premier Biosoft International, Palo Alto, CA) aid in both primer design and target sequence selection.
The QX200™ Droplet Digital PCR system provides absolute quantification of target DNA or RNA molecules with EvaGreen or probe-based digital PCR assays. The advantages of using hydrolysis probes include high specificity, a high signal-to-noise ratio, and the ability to perform multiplex reactions.
When designing probes, follow these guidelines:
- The Tm of each hydrolysis probe should be 5–10°C higher than that of the corresponding primers. In most cases, the probe should have <30 nucleotides between the fluorophore and the quencher to avoid affecting baseline signal intensity. It must not have a G at its 5' end, because this may quench the fluorescence signal even after hydrolysis. Choose a sequence within the target that has a GC content of 30–80% and design the probe to anneal to the strand that has more Gs than Cs (so the probe contains more Cs than Gs).
- Black Hole quenchers are recommended
- Tm enhancers for the probes are recommended for SNPs and rare event detection assays to minimize background fluorescence
The QX200 system is compatible with FAM dye and either HEX or VIC as a secondary dye. Using these dyes together in the same reaction, in which each is used to label a different target-specific hydrolysis probe, enables multiplex experiments and the quantification or detection of up to two targets per sample.
The quality of sample preparation can impact droplet digital PCR results. For best results observe the following guidelines:
- Use an optimized protocol to extract the DNA or RNA from the raw material that you are testing; use of crude lysates in the QX200 ddPCR system is not recommended
- Ensure the sample has not been degraded
- Although PCR inhibitors may be less detrimental to quantification accuracy when using ddPCR, care should be taken to remove as many of these as possible during the nucleic acid purification phase. If known inhibitors cannot be readily removed, consider diluting the sample 1:10. This dilutes the inhibitor tenfold, hopefully to a concentration that no longer impacts the PCR reaction
Sample concentration can also affect results. The recommended dynamic range of the QX200 ddPCR system is 1–100,000 copies per 20 µl reaction.
- If the experiment entails quantitating samples known to have extremely high amounts of target molecules (for example, NGS libraries), dilute the initial sample accordingly
- If the target copy number per genome is unknown, determine the optimum dilution by performing a three- or fourfold dilution series of each sample in the expected digital PCR range. Assaying three to four data points above and below the expected range ensures that one of the data points lies within the optimum digital range
To help determine copy number per genome, collect the following information:
- If the source or species of the gDNA is known but the genome size of the organism of interest is unknown, refer to http://www.cbs.dtu.dk/databases/DOGS/index.html to determine the size of the genome in question
- Once the size of the genome is known, determine the mass of the genome using the following formula:
m = n (1.096 × 10–21 g/bp)
where m is the genome mass in grams, and n is the genome size in base pairs.
The following example calculates the mass of the human genome using the Celera Genomics estimate of 3.0 × 109 bp (haploid):
m = (3.0 × 109 bp) (1.096 × 10–21 g/bp)
m = 3.3 × 1012 g or 3.3 pg
The example is relevant to any gene that is present at the usual frequency of two copies per diploid genome, such as RPP30, and provides a basis for performing a digital screening experiment to determine the optimal digital range.
For sample DNA loading, follow these guidelines:
- Assess input DNA/RNA concentration using NanoDrop (trademark of Thermo Fisher Scientific Group) or A260 spectroscopy to ensure that the target DNA/RNA concentration is within the dynamic range of detection. The accuracy of this measurement depends on a number of factors, including pH and the presence and concentration of contaminants, buffers, and salts, and therefore provides only an estimate of your sample DNA concentration
- It is recommended to run a maximum of 50 ng/µl digested or fragmented DNA (1 µg per 20 µl reaction)
- Intact DNA at concentrations above 3 ng/µl (60 ng per 20 µl reaction) requires restriction digestion to produce droplets of the correct size. For CNV, perform restriction digestion at all DNA concentrations unless access to proximal replicate sequences is necessary
- Do not restriction digest the DNA sample within the amplicon sequence
- Fragmented DNA such as that from formalin-fixed, paraffin-embedded (FFPE) tissue samples does not need restriction digestion