So you’ve got the results from your first qPCR experiment…well done!
The next step is to transform your raw data and write up the results as a report.
Our double delta Ct calculator turns your Ct results into a fold change and flags assumptions that might negatively affect it. Each flag this calculator raises links straight to the section that tells you what to do about it, so a suspicious result becomes a next step rather than a dead end. Ready? Let’s jump in!
Check your method first
There are two common ways to analyze qPCR data for relative expression: the double delta Ct method (Livak) and the efficiency-corrected Pfaffl model. A third, separate relative standard curve approach also exists (primarily for measuring amplification efficiency or estimating relative amounts) but it isn’t the same as the Pfaffl calculation. The practical decision here comes down to your amplification efficiencies.
Choose a free resource to help you move forward
REFERENCE CARD
qPCR Efficiency & Ct Reference Card
Troubleshooting Card
RT-qPCR Method Selection
| Method | Use when | What it does about efficiency |
|---|---|---|
| Double delta Ct / Livak (2−ΔΔCt) | Your target and reference assays amplify at similar efficiencies, both close to 100% (roughly 90–110%) | Assumes a perfect doubling every cycle for both genes |
| Pfaffl (efficiency-corrected) | Your target and reference efficiencies differ by more than a few percentage points, or sit outside ~90–110% | Uses each assay’s measured amplification efficiency instead of assuming 100% |
Double delta Ct analysis rests on three assumptions: your target and reference assays have comparable amplification efficiencies (comparable standard-curve slopes, typically within about 5 percentage points of each other), both amplify at close to 100% efficiency, and your internal control gene is expressed constantly and unaffected by the treatment.
If your measured efficiencies differ by more than that, the fold change drifts, and the drift grows with the size of your ΔCt. Check your amplification efficiency from a standard curve, and if the assays don’t match, switch to the Pfaffl method (or enter your efficiencies in the calculator below).
Before you start
These are the checks you should run every time you do a PCR. Skip them and your answer might be wrong!
- Average your technical replicates first. Feed one Ct per gene per sample into the method. Treat “wells agree within about 0.5 Ct” as a practical screening threshold to catch pipetting problems, not a hard validity criterion. Check your own platform’s tolerance, and if the spread is wider, fix it before going further.
- Validate your reference gene in your conditions. A housekeeping gene that’s stable in one cell type can shift under your treatment. Confirming that Ct barely moves between control and treated is a quick screen, not full validation: for publication-quality data, validate two or more candidate reference genes across your biological replicates and conditions.
- Use the same calibrator throughout. Whatever you pick as the baseline, keep it consistent across every sample and report it.
- Keep your Ct values in the reliable range. On most assays, Ct values above about 35 sit near the detection limit. Start from good-quality RNA and quality-control your samples so you’re not quantifying noise.
Double Delta Ct Calculator
Do you have your four averaged Ct values? Good! Here’s how you can calculate fold change and assess whether your result is trustworthy.
Enter your four average Ct values below. The calculator flags anything suspicious and shows you exactly where to look next. Leave the efficiency fields blank for the standard 2−ΔΔCt calculation, or enter both to switch to the efficiency-corrected Pfaffl model. A real assay should sit at 90–110% efficiency, and the calculator warns you if an entry falls outside that range.
qPCR Fold Change Calculator
Calculate relative gene expression (2−ΔΔCt) from your Ct values — with the full working shown step by step.
Wherever your result lands, read next: what your result means, my result looks wrong, and what the protocol doesn’t tell you.
The 4 steps behind the calculator
Here’s what the calculator just did, so you can reproduce it by hand or in Excel (for a detailed treatment, read the Livak paper). You need four averaged values: the gene being tested in the experimental (TE) and control (TC) conditions, and the housekeeping gene in the experimental (HE) and control (HC) conditions.
One note on terminology: you’ll also see Ct written as Cq (quantification cycle). The MIQE guidelines standardize on Cq, but we use the more familiar Ct throughout here. They refer to the same number: here’s what a Ct value is if you want the background.
| Step | Calculation | What it does | Worked example |
|---|---|---|---|
| 1. Average Ct values | Mean of technical replicates → TE, TC, HE, HC | Gives one Ct per gene per condition | TE 21.40, HE 18.20, TC 25.10, HC 18.30 |
| 2. ΔCt | TE − HE and TC − HC | Normalizes the target gene to the housekeeping gene | ΔCTE 3.20, ΔCTC 6.80 |
| 3. ΔΔCt | ΔCTE − ΔCTC | Compares experimental to control | ΔΔCt −3.60 |
| 4. Fold change | 2−ΔΔCt | Converts the log-scale difference back to a fold change | 12.13 |
The final step matters because everything is in base-2 logarithms, meaning each doubling of DNA lowers the Ct by 1, so a Ct difference doesn’t halve or double linearly. Raising 2 to the power of-ΔΔCt converts the log-scale difference back into a fold change you can report.
Here, the negative ΔΔCt gives a 12-fold upregulation, and the reference gene barely moved (18.30 versus 18.20), which is what a trustworthy run looks like. Enter these values into the calculator above to see the same result without flags.
A ΔΔCt is a difference of differences. A systematic error in any of the four Ct values it’s built from carries straight through to the fold change, so it’s worth getting it right first time.
What does your result mean?
Your fold change is the expression of your gene of interest in the test condition relative to the control, normalized to your housekeeping gene. Think of it as a percentage, e.g., a value of 1 means 100% as much expression as the control, so no change.
| Fold change | Meaning | What to check |
|---|---|---|
| > 1 (e.g. 5) | Upregulation (5 = 500% of control) | Confirm with replicates; is the effect biologically plausible? |
| ≈ 1 (roughly 0.8–1.25) | No large change flagged | Reference gene stability; confidence intervals and statistics before you call it either a change or no change |
| < 1 (e.g. 0.5) | Downregulation (0.5 = 50% of control) | That your calibrator and reference gene are correct |
A few things to keep in mind:
- A fold change near 1 isn’t automatically “no effect,” and a t-test on fold changes is the wrong test. Do your statistics on ΔCt values.
- The 0.8–1.25 band is a practical flag we use to prompt a second look, not a formal biological equivalence interval: whether a near-1 value means “no effect” depends on your assay variability, statistical power, and confidence intervals, and a well-powered study can report no meaningful change as a real finding.
Present the data as fold-change bar charts with the control set to 1. Run your statistics on ΔCt (or log-transformed) values with a test appropriate to your design: that means accounting for your biological replicate structure, using paired tests for matched samples, and correcting for multiple comparisons when you test several genes.
A simple t-test only fits a simple two-group, independent design. To reuse the workflow, build an Excel template once and just drop your data in each time. If the number doesn’t match your expectations, the troubleshooting section below covers the usual culprits.
My result looks wrong
Double-delta Ct analysis fails quietly. Use this as a troubleshooting checklist: each pattern below produces a fold change that appears to be a real result but isn’t.
Your fold change flips direction when you change nothing but the calibrator
Most likely cause: You’re computing ΔCt as reference minus target in some samples and target minus reference in others, which inverts the sign.
Try this: Fix the subtraction order to target minus reference (TE − HE) everywhere, and keep one calibrator across the whole experiment.
Every fold change looks off, even samples you expected to be flat
Most likely cause: Your housekeeping gene shifted between conditions, so the normalizer itself is moving. Under near-100% efficiency, a one-Ct shift in the reference is roughly a two-fold error carried into every sample; with efficiency-adjusted analysis, the exact size depends on your efficiencies.
Try this: Check that HE and HC agree within about 0.5 Ct as a quick screen. If they don’t, validate a more stable reference gene or normalize to the geometric mean of two or more validated reference genes (the geNorm approach).
A gene you know is expressed comes back at a Ct above 35, with a huge fold change
Most likely cause: At Ct > 35 on most assays, you may be detecting primer-dimers or background, not target. Small differences at the detection limit explode into large fold changes.
Try this: As a common threshold, treat Ct > 35 as not reliably quantified. Optimize the assay or increase input rather than trusting the number.
Replicates that should match give a wide spread of fold changes
Most likely cause: Pipetting error at the technical-replicate stage, amplified by the exponential nature of Ct.
Try this: Reduce pipetting error and tighten your replicates, and if the spread persists, trace the problem back to reverse transcription.
What the protocol doesn’t tell you
- Your choice of calibrator changes every number you report. In relative qPCR, every result is measured against a baseline sample set to 1. You can build that baseline in two ways (average all your control replicates or use a single control well), and each yields different fold-change values. Neither is wrong, but the two aren’t comparable, so pick one approach, use it across your entire dataset, and report it in your methods section.
- A “stable” housekeeping gene is only stable until your treatment indicates otherwise. GAPDH levels shift under low-oxygen conditions, and β-actin changes during differentiation. Since every fold change is measured relative to that reference, a housekeeping gene that doesn’t remain constant can distort all your results. So make sure you validate two or more candidates under your actual conditions (see MIQE and geNorm, refs 3 and 4) before trusting them.
- Do your statistics on ΔCt values, not on fold changes. Fold changes are exponential and heavily skewed, so averaging them or running a t-test on them is mathematically wrong. Run the test on ΔCt values (or log-transformed values), and then convert the mean back to a fold change for the figure.
- 2−ΔΔCt quietly assumes a perfect doubling every cycle. Only assays with an amplification efficiency in roughly the 90–110% range are considered acceptable, and real assays that dip below that need attention rather than trust. When your target and reference efficiencies differ by more than a few percentage points, the assumption bends the result more the larger your ΔCt gets. That’s the point to switch to the Pfaffl method, or to enter your efficiencies in the calculator above.
Common mistakes
| Mistake | How to spot it | How to prevent it |
|---|---|---|
| Reversing the subtraction order | Fold changes point the opposite way to your biology | Always ΔCt = target − reference (TE − HE) |
| Averaging fold changes across replicates | Means look inflated versus the individual values | Average ΔCt, then convert once to fold change |
| Trusting Ct values above 35 | Huge fold changes from low-abundance targets | Flag Ct > 35 as not reliably quantified on most assays |
| Changing calibrator mid-analysis | Samples stop being comparable | Pick one calibrator and keep it for the whole set |
| Ignoring a drifting reference gene | Even flat samples come back changed | Confirm HE and HC agree within ~0.5 Ct |
Further reading
Want to get extra credit? Deepen your understanding of what a Ct value actually is, the full picture of qPCR data analysis, and the essential qPCR papers worth reading. For more qPCR troubleshooting and the wider technique, see the RT-qPCR hub.
References
- Livak KJ, Schmittgen TD. Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2−ΔΔCt Method. Methods. 2001;25:402–8.
- Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001;29:e45.
- Bustin SA, et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem. 2009;55:611–22.
- Vandesompele J, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 2002;3:research0034.
You made it to the end—nice work! If you’re the kind of scientist who likes figuring things out without wasting half a day on trial and error, you’ll love our newsletter. Get 3 quick reads a week, packed with hard-won lab wisdom. Join FREE here.
