Documentation · Exports
Exports and the Methods section
Every export is built from the same stored analysis, so the numbers in a PDF, a spreadsheet and a Prism file are the ones on screen.
Export formats
| Format | Contains |
|---|---|
| PDF report | Results table, one figure per gene with its test and every comparison as a ratio with its CI, the MIQE 2.0 checklist, parameters, exclusions, and the two-way ANOVA of two-factor batches. |
| Excel (.xlsx) | A summary, the efficiencies, one sheet per gene that recomputes the calculation with live formulas from the raw Cq, a Statistics sheet laid out as Prism reports results, reference-gene stability and the two-way ANOVA when they apply, and the parameters. |
| CSV | One flat row per sample and gene, with every intermediate value from efficiency to fold change, and the pair ID or second factor when the design has one. |
| JSON | The audit trail: batch settings, efficiencies, every statistic exactly as the engine computed it, manual exclusions with their reasons, and every result row. |
| Prism (.pzfx) | Two column tables per gene, fold change and log2 fold change, ready for Prism’s column analyses. |
| Lab notebook (.zip) | The PDF, the Excel workbook and the JSON together, to attach to Labguru, Benchling, eLabFTW or RSpace. |
| Figures | Vector PDF, SVG, PNG and TIFF at the resolution you choose, one file per gene. Researcher plan. |
Estimates in the exports are on the scale the test ran on: on log2, a comparison’s estimate is a log2 ratio and the fold-change ratio is 2 raised to it. The Excel Statistics sheet and the PDF already show the ratio.
Opening in Prism
The Prism file holds, for each gene, a column table of fold changes and the same table on log2. Rows are titled with the sample, or with the pair ID in a paired design, where paired samples share a row as Prism expects. To reproduce Ampliova’s result in Prism, analyse the log2 table:
- Two groups: Unpaired t test with Welch’s correction (or Paired t test for a paired design).
- Three groups and up: Ordinary one-way ANOVA, comparing each column with the calibrator column by Dunnett’s test, or every pair by Tukey’s.
Prism then gives the same P values, and its differences of means on log2 are the log2 of Ampliova’s ratios. The fold-change table is there for the figure.
The Methods section
The Methods sub-tab of the Results tab lays out a Methods section as it will read in your paper: three bench sections you write (RNA extraction, reverse transcription, quantitative PCR) and a data-analysis paragraph Ampliova writes from what the analysis actually did: the quantification method and its references, the efficiencies, the outlier and replicate rules and how many wells each removed, your manual exclusions, the normalization, the calibrator, and the statistics: test, multiple-comparison procedure and family, scale, unit of replication, α, how effects are reported, the Benjamini-Hochberg adjustment on RT² arrays, and the versions of SciPy and NumPy the engine ran.
It is a template filled from the stored analysis, not generated text: the same batch always gives the same paragraph, and it states nothing the analysis did not do. It names MIQE 2.0 only for the efficiency correction and multi-gene normalization it applied, since compliance with the guideline depends on the whole experiment [2]. Researcher plan.
Everything is editable and saved with the batch as you type. The qPCR section starts from what the plate files recorded: the instrument, the cycling protocol when the RDML carries one, and the sequences of the primer pairs linked to the batch. If you edit the data-analysis paragraph and later run the analysis again, the tab says so and shows the new paragraph next to yours, so the text cannot drift away from the numbers without you knowing.
Save the bench sections as a preset to start the next batch from them. A preset never contains the data-analysis paragraph, which always comes from the batch it describes. On the Lab plan, share a preset with your team so everyone starts from the same protocol text; only its owner changes it. Copy the Methods section pastes into Word or Google Docs with its headings and the list of the works it cites.
What stays yours
Ampliova is never told how the RNA was extracted, checked or reverse transcribed, so it does not write those sections for you. The field hints list what MIQE 2.0 expects each one to state.
Citing Ampliova
In your Methods, name the software (“qPCR data were analysed with Ampliova, ampliova.com”) and cite the method it implements, Müller and Tiret [1]. Cite the underlying methods you relied on: LinRegPCR [3, 4], or the ΔΔCq method [5] if you used it; geometric-mean normalization [10]; the reference-gene measures if you report them; and MIQE 2.0 [2]. The generated paragraph includes these citations. The full list is on the references page.