FOR PHARMA & BIOTECH R&D
Right the first time. Comparable across every run.
In discovery, a qPCR result decides a target, a lead, a hit. Clarida runs your whole method, everything except the cycler, the same way every run, so the difference between two runs is the biology, not who analysed them.
When the wrong result steers a program
It looks fine until it is too late.
Efficiency left at 100%, a reference-gene panel from another model, an outlier dropped, and the number looks fine.
Treated vs control no longer line up, because each run was normalized differently and the readout never shows it.
Cq, then Excel, one person’s R script, Prism and a separate geNorm detour: method slips in at every handoff.
Six months on, is a discrepancy real biology or a different spreadsheet? And the sample may be gone.
The methods the field already trusts, every run.
Reference genes scored for this model, not carried over from another, so an unstable one is caught early.
Your lab authors the method once, and every scientist starts from it, not a blank spreadsheet.
The run stays on your own instrument, and your raw curves and Cq are never overwritten.
Efficiency-corrected quantification, with QC that surfaces outliers and drift instead of hiding them.
A self-contained, MIQE-aligned record that reopens with its data and methods, showing how a number was made.
Confidential by design
Private by default. Shared only when you choose.
Each scientist or group works in its own workspace, and workspace membership is the only way into a group’s experiments, samples, assays and results. That holds under one company account, and it holds for the account owner: no membership, no read.
So competitive programs down the hall run on one shared method without ever sharing their work with each other. What the account shares is the billing and capacity pool, and the system catalogs of instrument types and standard protocols every group can build on. Never a group’s data.
Data is encrypted in transit and at rest and hosted in the EU, so unpublished results never leave a group’s own workspace. When your IP and security team want the detail, we will walk them through exactly how it works.
The D-DEAR method
Right the first time, and comparable after.
One continuous workflow, from the question through the report, so the numbers a program compares stay comparable.
That’s D-DEAR: Define, Design, Execute, Analyze, Report. Everything except the cycler.
A discovery run frames its own question: a target-validation knockdown, a dose-response, a biomarker confirmation of an omics hit are not the same experiment, and the model they run in matters to how the number is read.
Because the context lives with the experiment, it travels to the record months later, when you or the scientist who inherited it needs to know why a run was set up the way it was, instead of reconstructing it from a lab notebook or losing it when a contractor rolls off.
Inside that preset, the assays carry their own acceptance criteria and the reference-gene panel is scored with geNorm, where you choose which genes to keep for this model. A scientist building a run pulls the preset instead of rebuilding a spreadsheet from memory.
From there each scientist designs the rest of their own experiment and keeps their own data private, so the lab shares a validated starting point without one rigid design across every question. A reference-gene panel validated in one model or tissue is not assumed to carry to another, so genes are re-scored when the biology changes.
Because the design is recorded, reference-gene choices, controls and acceptance criteria stay visible into every analysis, instead of becoming an undocumented source of variance that surfaces after a program has already moved.
A materials checklist, mixes at the calculated volumes, and a guided plate fill, then the run starts on your own instrument. Clarida guides and records here, it does not pipette or drive the cycler.
As a plate is loaded, anything visible on a well, a bubble, a short pipette, a cloudy reagent, can be noted, and that note stays attached to the well.
So in analysis the note sits next to the measurement, and an exclusion is one the scientist can explain rather than reconstruct, and a bench artifact does not get read as biology.
Bring raw amplification curves and Clarida can call Cq itself, with more than one established method to choose from, or start from the Cq values your instrument produced. The imported Cq is never overwritten.
From there: reference-gene stability scored with geNorm so you drop an unstable gene for this model, efficiency-corrected quantification (relative NRQ or standard-curve absolute), per-metric QC that surfaces outliers without dropping them for you, and inter-run calibration across plates, so treated vs control, dose 1 vs dose 8, this batch vs last month, and the qPCR readout vs the omics hit it confirms are actually comparable.
That comparability is the load-bearing part when a number decides a program: the difference between two runs is the biology, not which analyst ran which spreadsheet on which day.
It is also how you confirm a screen hit: when qPCR checks a differential expression from RNA-seq, microarray or a CRISPR screen, proper normalization and reference-gene scoring separate a real effect from a screen artifact. Clarida supports your call, it does not validate the biology for you.
When two results must be compared or a call is questioned, the exact analysis reopens: the method, the settings and the data behind every value, already on the record, so you do not re-run material you may not have.
The methods and settings are captured the way MIQE asks you to report them, so the record is complete and every result traces back to how it was produced.
The record carries its own data inside it, so it still works months later: it opens and re-analyses without Clarida, a database, or the software that made it. That self-contained, versioned record is the tie-breaker when two numbers disagree.
Why switch
One method every run, instead of a relay that drifts.
Right now each run is set up from scratch, and the analysis is a relay. One scientist decides their own plate layout, reference genes, replicates and controls, the instrument gives you Cq, then you export to Excel, someone runs a delta-delta-Cq calculation in an R or Python script, reference genes go through a separate geNorm or GenEx detour, and the figure is built in GraphPad Prism. Every handoff is a place for method to slip in, and it slips in differently each time, so two runs stop being comparable and the drift hides inside the fold-change.
Clarida does not replace your cycler or your ELN. It runs the method around the run: efficiency correction, normalization to reference genes you selected with geNorm, QC and inter-run calibration, applied the same way every run and recorded. One consistent method in place of a relay that drifts, so the runs a program compares are actually comparable.
Today, a tool for every step
Exported, reformatted, and pasted between apps that never talk.
With Clarida
- Definequestion & model
- Designyour lab’s method
- Executeyour cycler, Cq imported
- Analyzesame method, comparable runs
- Reporta record you can reopen
One record, from the question you framed to the figure you publish.
Right the first time
Efficiency-corrected quantification, reference genes you scored with geNorm, and QC that surfaces outliers instead of hiding them, so the number reflects the biology, not the method.
Comparable across every run
Inter-run calibration and the same method every run, so treated vs control, dose vs dose, and batch vs batch line up, and results stay comparable across projects and over time.
Re-scored, not carried over
A new cell line, tissue or model can break a reference-gene panel that worked before. geNorm re-scores stability for the new model, and you choose which genes to keep, so an unstable one never skews the result.
From the source
Trust the analysis, because they wrote it.
Clarida comes from the team behind geNorm, qbase+ and the MIQE guidelines the field reports to. The methods you rely on stay in the hands that defined them, now maintained and moved to the cloud.
28,000+
citations of the geNorm and qBase methods
Since 2007
building qPCR analysis software
MIQE
the reporting standard, co-authored by the team
For group and platform leads
One standard across every project.
The analyst is no longer the variable
Author the method once, and every scientist starts from it. Run-to-run difference is the biology, not who ran which spreadsheet, so a wrong go/no-go is not an artifact of how it was analysed.
One account, every group isolated
One company account and one budget cover every group, each in its own workspace that no other group, and no org admin, can read into. You bring groups on as they are ready, and capacity scales across all of them.
Comparable across projects and years
Inter-run calibration and one method mean decisions that compare one project to another, or this year to two years ago, are safe to make. The method is the constant, not the answer.
Reopen the analysis, do not re-run the material
When a call is questioned or two results disagree, the exact analysis reopens with its data and methods on the record, so you reconstruct how a number was made instead of re-spending a sample you may not have.
The method outlives the people
When a contractor or postdoc rolls off, the method and the analyses stay in the workspace, so the program’s work does not walk out the door with them.
Ready to bring your method into one place?
Book a demo, and we will shape the deployment and the pricing around the programs and groups you run.
Book a demoQuestions, answered
Access is gated by workspace membership, nothing else. To read a workspace, you have to be a member of it, so one team never sees another’s experiments or results, even under the same company account, and even the account owner has no read into a team’s data.
The same rule holds beyond your own company: another organisation on Clarida, including a competitor who also uses it, is simply not a member of your workspaces, so there is no path to your data. The only things shared across your own account are the billing pool and the system catalogs of instrument types and standard protocols. Never a team’s work.
Your data is encrypted in transit and at rest, and hosted in the EU.
Yes. Your data is yours, and you can export it whenever you want. Nothing is locked to us.
Every report is self-contained: it carries its own data and methods, and it reopens and re-analyses without Clarida, a database, or the software that made it. What you build stays usable on its own, not trapped in a format only we can read.
For a deployment of the right size, dedicated infrastructure, and running Clarida inside your own cloud tenancy, is on the table. It is arranged on request and shaped around your programs, not a standard tier you switch on.
The shared platform already isolates every workspace by membership and hosts data encrypted in the EU, so many teams start there and move to a dedicated environment only if their IP or security review requires it. Bring those requirements to the demo and we will map them to what is possible.
No, and we will not pretend otherwise. Clarida is a research-use analysis platform, built for discovery and preclinical R&D, not a validated GxP or Part 11 system.
The record it produces is defensible to scientific scrutiny, a colleague, a repeat, your future self, because it reopens with its data and methods. That is a research-use record, not a regulated one.
If your Excel-and-R pipeline is documented, reproducible, and comparable across every scientist and every run, keep it. In practice it is hard to prove it was applied identically each time, it drifts between runs, and it gets re-derived by hand every time the model changes.
Clarida is that pipeline as one recorded method: geNorm stability scoring, efficiency correction, normalization and inter-run calibration built in, so it survives the person who built it and the runs stay comparable without rebuilding it each time.
And you do not throw anything out. Clarida imports the data and the analyses you already have, so you build on your existing work rather than start from zero.
Clarida is instrument-agnostic and reads qPCR and RT-qPCR data from the major platforms, whichever cyclers you run. Start from the Cq values an instrument called, or bring the raw amplification curves and let Clarida call Cq itself, using single-threshold, SDM or LinRegPCR. Your imported Cq is never overwritten.
It also reads qbase+ files, so a group with legacy analyses can bring them across rather than start from zero.
Reference-gene stability is scored with geNorm so you can drop an unstable gene for a given model, or use global-mean normalization. Quantification is efficiency-corrected, relative results come back as a geometric-mean fold-change with confidence intervals and Benjamini-Hochberg across genes, and absolute quantification runs off a standard curve. The tests are the ones qPCR papers use: t-tests, Wilcoxon, Mann-Whitney, one-way ANOVA and Pearson or Spearman.
Inter-run calibration ties plates together so runs stay comparable. This is method consistency, not a promise of the same answer: different inputs give different results, the method applied to them is the constant. Every step is recorded and the report carries its own data and methods inside it, so you can reopen the exact analysis months later and re-analyse without Clarida. Nothing is a black box.
A group runs as one company account and one budget, with capacity that scales as you bring more groups on, and each group isolated in its own workspace.
The specifics are a short conversation shaped around your rollout, not a fixed tier you have to fit into. You can start free and try it before any of that.
Analysis today is qPCR and RT-qPCR, read from the major platforms. dPCR support is on the roadmap.
If you run dPCR alongside qPCR for CNV or rare-allele work, raise it in a demo and we will be straight with you on timing.
Make the next number one you can act on.
Try it free on your own runs, or book a demo for the whole group.