Blog
Back to Blog
June 9, 2026

Converting Rotor-Gene Data to RDML Format for Analysis

The Rotor-Gene Q (and its older siblings, the Rotor-Gene 3000/6000) exports data in its own .rex format by default, which is essentially unreadable by anything except the Rotor-Gene Q software itself. If you want to bring your data into a platform-agnostic analysis pipeline — or just share it with a collaborator who doesn't have a Rotor-Gene on their bench — you need to convert it to something more universal. RDML (Real-time PCR Data Markup Language) is that universal format, and the Rotor-Gene Q software can export it natively, though the process isn't exactly intuitive.

Here's the short version: Open your run in Rotor-Gene Q software (v2.1+), go to File → Save As → RDML File, and you'll get an .rdml file containing your raw fluorescence data, sample annotations, and Ct values. If you're on an older software version that doesn't support RDML export directly, you'll need to export as .csv and then use the RDML-Convert tool from the RDML consortium to build your RDML file manually. Both paths work, but the native export saves you significant headaches.

Why RDML and Why Bother

RDML is an open XML-based standard designed specifically for qPCR data exchange (Lefever et al., 2009, Nucleic Acids Research). It was developed precisely because every instrument manufacturer invented their own proprietary format, making cross-platform analysis a nightmare. An RDML file bundles together things that usually get separated during export: the amplification curve data, sample and target annotations, plate setup, and analysis settings.

This matters practically in a few scenarios:

Native RDML Export from Rotor-Gene Q Software

If you're running Rotor-Gene Q software version 2.1 or later (most current installations are on v2.3.x), RDML export is built in. Here's the step-by-step:

  1. Open your .rex file in the Rotor-Gene Q software.
  2. Complete your analysis first. Set your threshold, assign sample names and types (Unknown, Standard, NTC, NRT), and define your targets. RDML export captures whatever annotations exist at the time of export, so if you haven't labeled your samples, the RDML file will contain generic identifiers.
  3. Go to File → Export → RDML (the exact menu path varies slightly between v2.1 and v2.3, but it's under Export in all versions).
  4. Choose RDML version. If given the option, select RDML v1.2 or v1.3 — these are the most widely supported by downstream tools. RDML v2.0 exists but support is still patchy.
  5. Save the file. It'll have an .rdml extension.

A few things to watch for:

Sample type mapping. The Rotor-Gene Q software uses its own nomenclature for sample types (Unknown, Standard, Negative Control, etc.). These map to RDML sample types, but if you've left everything as "Unknown" — including your NTCs — that's what goes into the RDML file. Fix your annotations before export.

Multi-channel runs. If you ran a multiplex assay with, say, FAM and HEX channels, the RDML export should include data from all channels as separate targets. Verify this by opening the RDML file in an RDML-compatible viewer or editor (the RDML consortium provides a free online tool at rdml.org). I've seen cases where only the "active" channel at the time of export gets included — toggle through all channels and run the analysis on each before exporting.

Rotor-Gene Q's 72-well rotor layout. Unlike 96-well plate instruments, the Rotor-Gene Q uses a circular rotor. RDML handles this fine — the well positions are just identifiers, not physical coordinates — but if you're combining data with plate-based instruments downstream, be aware that well "1" on the rotor doesn't correspond to "A1" on a plate. This is a metadata issue, not a format issue, but it confuses people.

The CSV-to-RDML Conversion Path

If you're stuck on older Rotor-Gene software (pre-v2.1) or the Rotor-Gene 6000 software that predates QIAGEN's acquisition, you won't have native RDML export. In that case:

Step 1: Export your data as CSV. In the Rotor-Gene software, go to File → Export and choose the CSV/text option. You'll typically want to export two things separately:

Export the raw data with the "Full Data Export" option if available. This gives you a matrix of fluorescence values: rows are cycles (1–40 or however many you ran), columns are wells/samples.

Step 2: Use RDML-Convert or build the file manually. The RDML consortium maintains rdml.org, which includes an online editor/converter. You can:

This is tedious for large experiments. For a 72-well rotor with 3 targets, you're annotating 72 positions. But it works, and you only have to do it once per run.

Alternatively, if you're comfortable with Python, the rdmlpython library (maintained by the RDML consortium) lets you script the conversion. A typical workflow looks like:

import rdmlpython as rdml

# Create new RDML object
rd = rdml.Rdml()

# Add experiment, run, targets, samples...
exp = rd.new_experiment(id="MyExperiment")
run = exp.new_run(id="Run_001")

# Add your data programmatically from parsed CSV
# (see rdmlpython documentation for full API)

rd.save("output.rdml")

This is the better approach if you're regularly processing Rotor-Gene data and want reproducible conversion. Write the script once, adjust sample annotations per run, and batch-convert your backlog.

Common Issues After Conversion

Missing baseline correction. The Rotor-Gene Q applies its own baseline correction algorithm (dynamic tube normalization). When you export raw fluorescence data to CSV and rebuild as RDML, you may be exporting the corrected or uncorrected values depending on which export option you chose. If your downstream analysis tool applies its own baseline correction (LinRegPCR does, for example), you want the uncorrected raw fluorescence. Applying baseline correction twice will distort your curves and inflate your efficiency estimates.

Ct vs. Cq terminology. The Rotor-Gene software reports "Ct" (cycle threshold). RDML uses "Cq" (quantification cycle), which is the MIQE-recommended term. They mean the same thing. Some tools that import RDML will display "Cq" and your Rotor-Gene printout says "Ct" — don't let this confuse your PI during lab meeting.

Efficiency values not transferring. If you ran a standard curve on the Rotor-Gene and it calculated an efficiency of, say, 97%, that calculated efficiency is stored in the .rex file but may not transfer into the RDML export depending on your software version. Check the RDML file for the amplificationEfficiency field. If it's missing, you'll need to recalculate efficiency from the standard curve points in your downstream tool, or add it manually to the RDML file.

File size. RDML files from the Rotor-Gene Q are typically small (under 5 MB even for multi-channel runs) because a 72-well rotor generates less data than a 384-well plate. This is rarely an issue, but if you're merging multiple runs into a single RDML file for a large experiment, file size can grow.

Validating Your RDML File

Before you hand off the RDML file for downstream analysis, do a quick sanity check:

  1. Open it at rdml.org using the online RDML editor. Verify that sample names, target names, and sample types (NTC, standard, unknown) are correct.
  2. Spot-check a few Cq values against what the Rotor-Gene Q software reported. They should match exactly — if they don't, you may have exported the wrong analysis or the threshold wasn't set when you exported.
  3. Check that amplification curve data is present, not just the Cq calls. Open the RDML file (it's just a renamed ZIP containing XML — you can change the extension to .zip and peek inside) and confirm the <adp> (amplification data point) elements contain your cycle-by-cycle fluorescence.

If the curves are there and the annotations are right, your RDML file is ready for analysis. VoilaPCR imports RDML files directly and automatically maps your samples, targets, and controls — so you can go from Rotor-Gene export to analyzed results without manually rebuilding your plate layout in a spreadsheet.

Related Posts