Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
A biomarker is only as good as the method used to quantify it. For protein and small-molecule biomarkers, researchers typically measure abundance using ELISA or mass spectrometry. Researchers interested in quantifying nucleic acids can try next-generation DNA sequencing, but more often they turn to some form of polymerase chain reaction (PCR), either real-time quantitative PCR (qPCR) or digital PCR (dPCR).
Though both methods are quantitative, they use very different hardware and protocols, and they yield data of different types. All things being equal, researchers can in most cases use either approach. But some occasions favor one option over the other. Here’s what you need to know.
Different types of PCR
The polymerase chain reaction works by amplifying a segment of DNA bracketed by two inward-facing oligonucleotide primers in multiple cycles of template denaturation, primer hybridization and polymerization. In theory, the amount of DNA in the reaction doubles with every cycle, so it should be possible to deduce the amount of template DNA at the start of the reaction by how much is present at the end of reaction—a strategy called endpoint PCR. But in practice, that isn’t always true.
That’s not to say the standard method is useless. “Endpoint PCR is still the workhorse of molecular biology,” says Gabriela Saldanha, strategic marketing manager for genomics products at Promega. Among other things, endpoint PCR is commonly used for cloning and genotyping applications and for answering quick yes/no questions.
Endpoint PCR can also be made “semiquantitative,” Saldanha notes, by running the products of the reaction on a gel alongside known quantities of DNA.
For more rigorous quantitation, researchers use either qPCR or dPCR. In qPCR, the reaction is monitored throughout the amplification process using fluorescent probes that increase in abundance (and hence intensity) with the amplification product. By recording the number of cycles required for the fluorescence intensity to rise above some threshold value and comparing that value (called Ct) to a standard curve, researchers can deduce the amount of DNA in the original sample.
qPCR equipment is widely available, including from Agilent Technologies, Analytik Jena, Bio-Rad Laboratories, Qiagen, Roche Life Sciences and Thermo Fisher Scientific.
Commercialized by Bio-Rad, Fluidigm, RainDance Technologies and Thermo Fisher Scientific, dPCR is essentially a cross between a limiting dilution experiment and endpoint PCR. Here, the template DNA, primers, fluorophores and other reagents are distributed into thousands of individual reaction chambers, such that, on average, each partition contains zero or one copy of the template. Following amplification, the system counts the number of positive partitions, applies a Poisson correction to adjust for the possibility of multiple template copies and then reports the absolute number of molecules in the original sample.
Pros and cons
Though both qPCR and dPCR are quantitative, they differ in the nature of the data they report. qPCR yields relative information—for instance, that a drug-treated sample contained twice as much of a given mRNA as the control. To determine the absolute number of molecules those signals correspond to, researchers must compare a sample’s Ct values to a standard curve. In contrast, dPCR provides absolute quantitation, no standard curve required. “The number of positive droplets gives a readout of the number of starting molecules,” explains Muneesh Tewari, associate professor of internal medicine at the University of Michigan Medical School.
qPCR and dPCR can also differ in how reliable those numbers are. Tewari, for instance, is investigating microRNAs circulating in blood plasma or serum for their value as cancer biomarkers. In such a situation, he says, reproducibility is key: Clinical laboratories must be able to accurately and reproducibly quantify biomarkers in samples collected over time and to compare those results to each other, for instance, to monitor a response to drug treatment. “We found that such reproducible absolute quantification was really not possible with real-time PCR,” he says. Although the method can be reproducible on the same plate or over the course of the same day, the absolute concentrations of microRNAs recorded day after day tended to vary by 2-fold or more in his hands.
In 2013, Tewari and his team published a head-to-head comparison of qPCR and dPCR—specifically, droplet digital PCR (ddPCR), a technology commercialized by Bio-Rad [1]. The two methods were more or less equivalent in terms of sensitivity, Tewari found. But ddPCR was considerably more precise and reproducible, reducing coefficients of variance by up to 86%.
dPCR may also be more resistant to PCR inhibitors, Tewari says, as dPCR reactions are read upon completion as a binary assay—either a partition was positive, or it wasn’t. Such data are less likely to be influenced by inhibitors, Tewari says, than in qPCR, where what matters is precisely when the reaction turns positive.
On the other hand, dPCR reactions tend to be more expensive than qPCR, and they require more hands-on time, Tewari says.
According to David Keys, assistant director of applications research and development in the Life Sciences Group at Thermo Fisher Scientific, dPCR reactions offer a relatively restricted linear dynamic range compared with qPCR. Thus, dPCR users typically must quantify their template prior to starting a reaction to ensure they fall within the linear range of the assay.
“If I have samples where I don’t know what my incoming concentration is, qPCR, with its very wide dynamic range, is a better choice,” Keys says.
Another difference between the two methods, he adds, is in their precision. qPCR can distinguish, say, a genome with the normal two copies of a gene from genomes with duplication leading to three or four copies. But if you need to differentiate seven copies from eight—a difference of just 15%—“that’s an example where we would advise looking at the dPCR approach.”
Saldanha recommends dPCR “for applications that are looking at more rare allele events, liquid biopsies and noninvasive prenatal diagnostics, where sample is very limited.”
Making a choice
In many cases, the choice between qPCR and dPCR comes down to personal preference. One limiting factor, though, is accessibility—researchers are more likely to have access to a qPCR instrument than dPCR, and they also tend to be more familiar with the method.
Certainly, qPCR users benefit from the extraordinarily rich collections of pre-designed assays that have been developed over the years. But according to Keys, these generally can be applied to dPCR with little or no modification.
As for Tewari, his 2013 study hasn’t made him give up on qPCR—he uses both techniques. “Where we need very precise measurements, especially those we want to compare day to day, then we use dPCR…. In other cases, [if] precision doesn’t matter that much or if the comparisons are all on the same plate, then we use qPCR, because the workflow is easier and less expensive.”
Reference
[1] Hindson, CM, et al., “Absolute quantification by droplet digital PCR versus analog real-time PCR, Nature Methods, 10:1003-5, 2013. [PMID: 23995387]
Image: Shutterstock images