Making Sense of qPCR Data: A Practical Guide to Analysis

Start with quality control

Before calculating any expression changes, it is worth checking whether the assay actually behaved as you expected.

Technical replicates should be reasonably consistent (hopefully you ran technical replicates!). Large differences between replicate Ct values can indicate pipetting errors, poor sample quality, or inconsistent amplification. We usually look for less than 0.5 Ct variation between technical replicates. No-template controls should remain negative, while melt curve analysis (for SYBR Green assays) should confirm that only a single specific product was amplified. Get in the habit of routinely checking the melt curves on every run. Amplification plots are also worth examining rather than relying entirely on the software to interpret these and output your Ct values. Irregular curves, unusually early or late amplification, or evidence of primer-dimer formation can all affect downstream calculations. All the curves from a particular primer set should have similar slopes or gradients. Variations in the slope of an amplification curve can indicate variations in reaction efficiency, which will directly impact the resulting Ct value, and ultimately the calculated fold change value. Contaminants in an RNA/cDNA sample may poison the PCR reaction and reduce its efficiency, producing amplification curves with a shallower gradient. These checks may seem tedious, but they often prevent hours of analysis on data that should not be used in the first place. See our melt curve analysis guide for SYBR Green specificity checks.

Understanding the Ct value

The quantification threshold value (Ct, sometimes called Cq) is the cycle at which fluorescence rises above background levels and crosses a specified threshold value. Lower Ct values indicate that more target nucleic acid was present at the start of the reaction, while higher Ct values indicate less starting material. However, Ct values themselves are not measurements of expression. They are simply the starting point for further analysis. Comparing raw Ct values between samples without normalization is rarely appropriate because differences may reflect variation in RNA quantity, RNA quality, or reverse transcription efficiency rather than true biological changes.

Why normalization matters

Normalization is one of the most important steps in qPCR analysis. The goal is to account for technical variation between samples so that observed differences are more likely to reflect genuine biological changes.

The most common approach is to normalize the target gene against one or more reference genes. These genes should be stably expressed across all experimental conditions. A common mistake is to assume that traditional housekeeping genes, such as GAPDH, are always stable. Numerous studies, including those discussed by Bustin and colleagues, have shown that reference genes must be validated for each experimental system rather than selected by habit.

Using unstable reference genes can introduce larger errors than the biological effects being measured. An alternative method is to normalise to the total amount of RNA added to the RT reaction. However, this requires very accurate quantification of RNA input amounts, assessment of RNA quality, using something like an Agilent Bioanalyzer, and confirmation that the RNA is DNA free. If you use total RNA normalisation you will probably also want to check out some housekeeper genes to establish that any observed change in your gene of interest is not simply due to a global change in transcription. Total RNA normalisation has been described as the least problematic normalisation method. For example, Stephen Bustin says “Normalization to total RNA is the least assumption-dependent and often the most reliable method, provided the quality and quantity of RNA are accurately assessed. Normalization to single ‘housekeeping’ genes can be misleading because their expression may vary under different conditions.” (Bustin 2002) and “Normalization against RNA concentration is the simplest solution…” (Nolan 2006).

Relative quantification and the ΔΔCt method

For many gene expression experiments, relative quantification is the preferred approach.

The workflow is simple:

1. Calculate ΔCt by subtracting the reference gene Ct from the target gene Ct.
2. Calculate ΔΔCt by comparing each sample’s ΔCt with that of a control group.
3. Convert the result into fold change.

This method is widely used because it is straightforward and works well when amplification efficiencies are close to ideal and similar between target and reference assays. For example, a fold change of 2 indicates approximately double the expression relative to the control, while a fold change of 0.5 indicates roughly half the expression.

Don’t ignore amplification efficiency

One of the recurring themes in the qPCR literature is that amplification efficiency matters. The traditional ΔΔCt approach assumes that amplification efficiency is effectively the same across assays (i.e. each reaction being 100% efficient). In practice, this assumption is almost never true. Small differences in efficiency can lead to substantial errors, particularly when comparing genes with very different amplification characteristics. This is why many researchers generate standard curves using serial dilutions. By plotting Ct values against template concentration, efficiency can be estimated and assay performance evaluated. Ideally, efficiencies should be close to 100%, meaning the amount of product approximately doubles with each cycle. Assays that fall far outside the acceptable range may require redesign or optimization.

Methods such as the Pfaffl model explicitly incorporate amplification efficiency into relative quantification calculations and can improve accuracy when efficiencies differ between assays.

The Pfaffl equation (also called the Pfaffl method for relative quantification in qPCR) accounts for differences in PCR amplification efficiency between target and reference genes:

R = (E_target)^(ΔCq_target(control−sample)) / (E_ref)^(ΔCq_ref(control−sample))

where:
  R = expression ratio (relative expression)
  E_target = amplification efficiency of the target gene
  E_ref = amplification efficiency of the reference gene
  ΔCq_target = Cq_control − Cq_sample
  ΔCq_ref = Cq_control − Cq_sample

If efficiencies are identical and equal to 2 (100% efficiency), the equation simplifies to the familiar 2^−ΔΔCq method.

Interpreting fold changes carefully

A common trap is assuming that any observed fold change is biologically meaningful. Recent work involving Stephen Bustin has highlighted how measurement uncertainty becomes increasingly important at low target concentrations. At very low copy numbers, technical variation can be large enough to rival or exceed the apparent biological difference between groups. In these situations, a reported two-fold change may not be as convincing as it initially appears. For this reason, fold changes should always be interpreted alongside measures of variability, confidence intervals, and appropriate statistical testing.

Statistical analysis is not optional

Once normalized expression values have been calculated, statistical analysis is needed to determine whether differences between groups are likely to be real. The exact test depends on the experimental design and distribution of the data, but the underlying principle remains the same: biological conclusions should not be based solely on fold-change values. Bustin and colleagues have repeatedly argued that qPCR should be treated as a quantitative measurement technique rather than a simple validation tool. That means reporting variability, describing analysis methods clearly, and avoiding overinterpretation of small differences. Also, remember that fold change values are not normally distributed (e.g. think of the difference between 1 and 10 vs 0.1 and 1). This means that fold change values should be log transformed prior to statistical analysis. Of course, if you do a log₂ transformation, this returns your fold change value back to the ΔCt value. qPCR is a log₂ assay after all!

Reporting your analysis

Even well-performed experiments lose value if the analysis is poorly documented. The MIQE guidelines were developed largely because many published qPCR studies failed to provide enough information for readers to evaluate or reproduce the work. Important details include primer sequences, assay efficiencies, reference gene validation, normalization methods, and statistical approaches. Transparent reporting helps other researchers judge the reliability of the findings and reproduce the experiment if needed.

Final thoughts

Good qPCR analysis is about much more than plugging Ct values into a spreadsheet. Reliable results come from careful quality control, appropriate normalization, consideration of amplification efficiency, and honest assessment of experimental uncertainty.

The central message from Stephen Bustin’s publications over the past two decades is surprisingly simple: qPCR is a powerful quantitative method, but only when data quality and analysis are treated with the same level of care as the laboratory work itself. Following those principles makes it far more likely that the biological conclusions drawn from a qPCR experiment will stand up to scrutiny.

References and further reading

• Bustin SA (2002), “Quantification of mRNA using real-time RT-PCR,” Journal of Molecular Endocrinology, 29:23–39. doi:10.1677/jme.0.0290023
• Bustin SA et al. (2009) The MIQE Guidelines: Minimum Information for Publication of Quantitative Real-Time PCR Experiments. Clinical Chemistry 55(4):611–622. doi:10.1373/clinchem.2008.112797
• Nolan, T., Hands, R. E., & Bustin, S. A. (2006). Quantification of mRNA using real-time RT-PCR. Nature Protocols, 1(3), 1559–1582. https://doi.org/10.1038/nprot.2006.236
• Pfaffl MW (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Research. 29(9):e45. doi:10.1093/nar/29.9.e45