2026.9·Make the call

Two months in one release. We didn't skip the August release for a summer break but because the raw data import, visualization and analysis only make sense as a package, and shipping them piecemeal would have meant interim releases where nothing quite worked end to end.

Until now, getting qPCR data into Clarida meant opening your instrument's software, exporting a table, and importing that table, with the raw curves behind those numbers staying behind on the instrument PC. This release closes that gap: Clarida fetches the file, reads the instrument's own format, shows you the curves, and, if you want, calls Cq itself. From raw instrument data to final interpretation, this release gets you there with full control at every step in between.

The second thread is about not doing the same setup twice. Kits and presets now carry their own analysis configuration, so a well-configured experiment can be started from a template rather than assembled by hand.

Reach the data where it lives

Your instrument PC is usually not the machine you analyze on. Clarida now runs a small on-site connector, called Porta, that makes those files reachable without a USB stick or a shared drive you had to talk IT into. Point it at a folder and that folder becomes an import source, wherever you happen to be working from.

Maximize

Porta runs on the instrument PC, not in the cloud, and only serves the folders you explicitly point it at. For a LightCycler 480 the connector goes one step further and talks to the instrument's own database, so you can browse and import runs directly rather than hunting for exported files. Further details on what it does, how it is set up, and how the security model works, are availiable in Connecting instruments with Porta in our methods library.

Import the instrument's own file

No more opening the vendor software just to produce an Excel export. Clarida reads the instrument file itself, for all four of the major platforms:

• Bio-Rad CFX · .pcrd
• Thermo Fisher / Applied Biosystems · .eds
• Roche LightCycler 480 · .ixo and .xml
• QIAGEN Rotor-Gene · .rex and .qrex

These join the RDML and qbase+ paths we already supported, and all of them stage through the same import flow, so run linking, plate layout, and sample and assay matching work identically no matter which file you started from. The raw amplification and melt curves ride along with the run, which is what makes the next two features possible.

Maximize

The file picker above shows both halves of this release's import story in one place: which formats Clarida can read, and where it can read them from. For a full account of what the importer reads and how the import steps adapt to your experiment, see Importing instrument data.

See the curves, not just the numbers

A Cq is a summary. When a result looks wrong, the curve is where the answer usually is, and until now that curve stayed in the instrument software. The Raw data section shows your amplification and melt curves inside the experiment, so you can slice them by run, sample, assay, or target, group them the way you actually think about the plate, and judge data quality before you trust a single number.

Maximize

A dedicated Raw data QC section sits alongside it, so curve-level problems surface as findings rather than as something you have to spot by eye.

Call Cq yourself, three ways

Use the Cq values your instrument calculated, exactly as before. Or have Clarida calculate them, independently of the vendor's software, with the method you consider appropriate:

• Single threshold · A fixed fluorescence threshold, the classic approach.
• SDM · Second-derivative maximum, threshold-free.
• LinRegPCR · Per-well efficiency from the log-linear phase.

Both sets of values live side by side. The well table in the screenshot above carries an Imported Cq column and a Calculated Cq column, so you can see immediately where the two agree and where they part company, rather than having to choose one and hope.

This matters most when you are comparing runs from different instruments, or when a reviewer asks how a Cq was derived. Every value carries its provenance: whether it came from the instrument or from Clarida, and by which method.

Kits that bring their own settings

A kit is more than a list of assays. The presets and kits in your library now carry their analysis configuration with them, including per-target detection thresholds, so starting an experiment from a kit gives you an experiment that is already configured, not just populated.

Maximize

Kit vendors can publish a kit to a named list of recipient workspaces, keep it private, or make it public, and change their mind later without breaking the experiments already built from it. Catalog numbers, protocol PDFs, and product links travel with the kit too, so the preset in your library points back at the product on your bench. If you run the same panel every week, this is the difference between setting it up every time and setting it up once.

Detection: present, absent, or nothing to say

For presence/absence work, a table of Cq values is not the answer. The call is. Detection reads your Cq data against a cutoff per target and gives you a sample-by-assay grid of calls, with wells that were never measured shown as exactly that rather than quietly counted as negative.

Maximize

Cutoffs come from the assay in your library, which is how a kit can ship its own, with an experiment-level default behind them and a per-target override in front. Your positive and negative controls are evaluated alongside the calls, so a control problem is visible in the same place as the result it undermines. Note the QC card in the screenshot: one sample out of seven is failing, not all seven. Control problems are attributed to the samples they actually implicate.

Import from your ELN

The electronic lab notebook attached to an experiment used to be something Clarida carried but never read. Now it is a source you can import from: samples, assays, and Cq values can all come out of the linked notebook, the same way they would from a file.

Maximize

Clarida only offers the notebook once it holds something worth reading: the stock template every experiment starts with has nothing to import, so the ELN appears as a source only after you have attached a workbook of your own.

• Aligned storage library · Presets, mixes, cycling protocols, instruments, and connectors now share the same layout, toolbar, and detail panel as samples and assays. Everything in Storage behaves the same way, including create and edit flows that no longer lose your work if you navigate away.
• Sortable, filterable tables everywhere · Every table across the qPCR workflow now shares one column header with sort and filter, drag-to-resize columns, freeze leading columns in place, and reorder columns by drag or menu.
• Smarter run linking · When you import a run into an experiment that already has runs, Clarida proposes the matching and explains why, instead of leaving you to line them up by hand. Run names can be edited in place when the proposal needs a nudge.
• The importer asks better questions · Unlisted dyes, implausible Cq values, and ambiguous columns now surface as specific, answerable questions during import rather than as a failure afterwards.
• Decimal entry that behaves · Comma decimals are accepted throughout, number fields no longer commit half-typed values while you are still typing, and every number in the app is formatted by the same rules. The full rule, including the one case Clarida deliberately refuses to guess at, is written up in Decimal separators.
• Data health at a glance · Exploration carries a roll-up of your QC status across sections, with per-section status lines for efficiency and normalization, an explicit acknowledgment when you have reviewed a deviation, and a checkpoint button to snapshot where you are.
• Large experiments load · Experiments with many samples and assays no longer time out on open or export as ELN.
• Faster updates · Changing a rescaling reference or editing group membership recalculates just what changed instead of re-running the whole pipeline. We will continue these optimization efforts to get you the speed and responsiveness you deserve.
• Clearer statistics sections · Consistent composition across the statistical tests, chart types that are honored in paired tests, a named rescaling reference instead of "None", and group brackets that no longer collide with column labels.
• New chart types: box plot, violin, dot strip · Statistics sections and Sample & Assay QC can now show the shape of your data, not just a bar and its error whiskers.
• Named QC ranges and control labels · Control rows are labelled by their assay-part, and a positive control with no specification is handled as its own case rather than as a failure.
• Import fidelity · RDML quantities now split fold-change from dilution correctly, exclusions recorded upstream survive the import, and tabular Cq values are interpreted consistently no matter which path they came in on.

This release fixes four things that were producing wrong numbers or wrong verdicts. If you ran and reported analyses before upgrading, these are worth a second look.