Getting started

What Ampliova does

An analysis is a batch: one experiment, with its target genes, its reference genes and a calibrator group, the condition every fold change is expressed against. A batch holds any number of plates and is always analysed as a whole, so the efficiency of each gene is estimated from all of its wells in the batch, whatever plate they sit on.

The method is the efficiency-corrected relative quantification described by Müller and Tiret [1]. LinRegPCR measures the efficiency of every well from its own amplification curve [3, 4], each gene gets the mean of its wells, and each sample is normalized to the geometric mean of at least two reference genes [10]. No standard curve and no dilution series are needed. The classic ΔΔCq method, which assumes a perfect doubling every cycle, stays available.

From plate to figure

1. Create a batch. From the dashboard, click New batch. Name it, list the target and reference genes, name the calibrator group, and choose the method: efficiency-corrected (the default) or classic ΔΔCq. This is also where you pick the batch type and, for expression batches, a two-factor design.
2. Add plates. Upload one RDML file per plate. Ampliova reads the Cq values, the amplification curves and the sample and target names in the file, and fills in the plate layout from them. Unless you or your lab turned them off in Settings, AI suggestions then propose groups and well types from the sample names, which are sent to Mistral AI for that and nothing else; every suggestion waits for your review. Check every well’s type (sample, reference gene, NTC, RT−), sample name and group. In a paired design, the confirmation step is where each sample gets its pair ID. That step also checks for a no-template control, for at least two reference genes, for controls that amplified too close to their samples, and for sample names that differ from other plates only by case or spacing.
3. Reuse a layout. If you run the same plate layout again and again, save it with Save as template on the Layout step, and apply it to the next plate with Apply a template: each well takes its type, gene, group and, if you want, sample name from the template, and keeps its Cq values and curves from the file. Wells are matched by position, and anything that does not fit (a template well with no reaction in the file, a gene that differs) is listed for you to check. Templates live under Dashboard → Plate templates, where you can also build one or import a layout from the free plate designer. Applying a template copies it into the plate, so editing the template later never changes an analysis. Researcher plan; on the Lab plan, share a template with your team: its members apply or duplicate it, and only its owner edits it.
4. Run the analysis. Once at least one plate is validated, Run analysis in the Plates tab computes the whole batch. The calculation takes seconds.
5. Read the results. Figures shows one chart per gene, with the test, its P value and each comparison as a ratio with its 95% confidence interval underneath. Table shows the full calculation chain for every sample, from Cq to fold change; click a sample or a single well to exclude it, then recalculate. Assumptions shows the residuals the test assumed normal, and lets you change the test.
6. Export. A PDF report with its MIQE checklist, an Excel workbook, CSV, the JSON audit trail, a Prism file, a lab-notebook bundle, publication figures, and a Methods paragraph written from what actually ran. See Exports and Methods.

Batch types and designs

An expression batch has one of three designs, and the design, not the data, decides which test runs. Unpaired is the default: each biological sample belongs to one group. Paired applies when the same animal, donor or culture is measured in every condition, and it is declared by giving those samples a common pair ID; Ampliova never pairs samples on its own. Two-factor crosses two conditions, such as genotype and treatment, with a second group field on each well. The statistics page lists the test each design gets.