Methods

The comparative Ct method, as described by Livak & Schmittgen (2001).

This method assumes approximately equal amplification efficiency (~100%) for both target and reference genes. If efficiencies differ substantially, use the Pfaffl method instead.

Statistical testing

For each target gene, VoilaPCR compares ΔCt values between each treatment group and the control group using a Welch's unpaired t-test (two-tailed). Welch's test does not assume equal variance between groups. P-values are computed from the t-distribution using the regularized incomplete beta function with Welch–Satterthwaite degrees of freedom.

Significance thresholds: *** p < 0.001, ** p < 0.01, * p < 0.05, ns = not significant.

Error bars on fold-change plots represent the standard error of the mean (SEM) of ΔCt values, propagated through the 2^−ΔΔCt transformation.

Efficiency-corrected relative quantification, as described by Pfaffl (2001).

Unlike ΔΔCt, the Pfaffl method accounts for differences in PCR amplification efficiency between target and reference genes. Efficiency values (E) can be entered manually or derived from a standard curve. Default efficiency is 2.0 (100%).

Statistical testing follows the same Welch's t-test approach as ΔΔCt, applied to efficiency-corrected ratios.

Multi-reference-gene stability assessment, implementing the algorithm described by Vandesompele et al. (2002).

M-value calculation

For each candidate reference gene, VoilaPCR calculates pairwise variation against every other candidate across all samples:

Lower M-values indicate more stable expression. VoilaPCR ranks all candidate genes by M-value and recommends the two most stable genes for normalization.

Normalization factor

The normalization factor for each sample is the geometric mean of the relative quantities (2^−Ct) of the recommended reference genes, following the original geNorm protocol.

Requires a minimum of 2 candidate reference genes and 2 samples common to all candidates.

Absolute quantification from a dilution series using linear regression of Ct versus log₁₀(quantity).

Acceptable efficiency range: 90–110%. R² must be ≥ 0.98 for reliable quantification. Both values are reported and checked by the automated QC system.

If a reference gene is provided, quantities are normalized: normalized quantity = quantity / reference quantity.

Technical replicates are automatically averaged before analysis. An outlier detection step flags individual replicates that deviate from the group mean:

When outliers are detected and at least one non-outlier replicate remains, the outlier is excluded from the average. If all replicates are flagged (e.g., high scatter across the entire group), all values are retained and averaged. Excluded wells and their deviations are reported in the QC output.

Every analysis runs 8 diagnostic checks automatically. Each check is scored as Critical, Warning, Caution, or Pass.

Overall QC score

The report is scored as Red (any critical issue), Yellow (warnings but no critical issues), or Green (all checks pass or cautions only). Every flagged issue includes an explanation and actionable suggestions.

VoilaPCR output has been validated against independently calculated results (manual spreadsheet calculations and qbase+) across multiple datasets. ΔCt, ΔΔCt, and fold-change values were confirmed to be identical to manually computed values.

All statistical functions (Welch's t-test, standard deviation, SEM) use unbiased estimators (n−1 for sample variance) and compute exact p-values from the t-distribution — not normal approximations.

All qPCR data processing — file parsing, Ct averaging, normalization, statistical testing, and figure generation — runs entirely in your browser using client-side JavaScript. Your spreadsheet data is never transmitted to any server. There is no server endpoint that accepts experimental data.