# Experiment settings sets

> How to derive your lab's settings set from Clarida defaults and change it in use, and how a set inherits and locks its values.

Source: https://clarida.bio/methods/settings-sets
Updated: 2026-11-10

[Methods](https://clarida.bio/methods)

How to derive your lab's settings set from Clarida defaults and change it in use, and how a set inherits and locks its values.

By [Jan Hellemans](https://clarida.bio/methods/settings-sets#about-the-authors)

October 12, 2026 Updated November 10, 2026 13 min read

Every qPCR analysis rests on choices made before the first number is computed. How far apart may technical replicates be before the set is flagged? Where does a target's amplification efficiency come from? Above which geNorm M value is a reference gene unstable? Is the plate a 96-well or a 384-well plate? A lab tends to give the same answers on every run.

An **experiment settings set** holds those answers once, as a named and versioned item in the Library. Every experiment started from a preset that carries the set begins with exactly its values, and records which set and version it started from.

**How to** covers deriving your own set, using it, and changing it later; **How it works** explains what a set holds and the rules behind it, for when a step raises a question. The running example is a lab that types SNPs by TaqMan allelic discrimination on 384-well plates. It wants a stricter replicate check than Clarida's default, and wants import to lead straight to the genotype table.

## How to

### 1. Look at what Clarida defaults holds

In the sidebar, open **Setup → Experiment settings** and select **Clarida defaults**. Its details list the presets and experiments in your workspace that use it. **View settings** opens the whole set, read-only:

- **The table** has one row per setting, with its **Group**, the **Application** it affects, the setting it depends on under **Conditional**, and its **Value**. The question mark beside a setting's name explains it.
- **Search settings…** finds a setting by its name or by words in its explanation.
- **Only settings that apply now** hides the settings that have no effect with the set's current choices.
- **Compare with…** places another set beside this one.

![The read-only view of Clarida defaults v2. The Details card gives its name and description: Clarida's recommended analysis, QC, design and workflow settings, the set experiments start from unless their preset names another. The Settings card has a Compare with… picker, an Only settings that apply now switch and a search box holding 'threshold'. The table lists the ten matching settings, among them ΔCq threshold 0.5 and SD threshold 0.25 in the QC group for Genotyping and RQ, the reference gene CV and geNorm M thresholds for RQ, and three inter-run calibration thresholds whose Conditional column names Inter-run calibration mode; one of them reads Same as ΔCq threshold](https://clarida.bio/images/methods/settings-sets-view-v2.png)

Figure 1: Clarida defaults, searched for 'threshold'. Three of the settings found apply only when inter-run calibration is applied.

See [What a settings set is](https://clarida.bio/methods/settings-sets#what-a-settings-set-is), [What a set holds](https://clarida.bio/methods/settings-sets#what-a-set-holds), [Conditional settings](https://clarida.bio/methods/settings-sets#conditional-settings), [Clarida defaults](https://clarida.bio/methods/settings-sets#clarida-defaults) and [Comparing two sets](https://clarida.bio/methods/settings-sets#comparing-two-sets).

### 2. Derive your own set

With Clarida defaults selected, choose **Duplicate**. (**New settings set** does the same, starting from the latest Clarida defaults unless you choose another set under **Start from**.) Every setting starts at its parent's value, so change only what your lab does differently:

- **Name the set.** The lab calls it **Genotyping 384-well**.
- **Change the values.** The lab sets **Plate format** to **384-well**, **ΔCq threshold** to 0.3 instead of 0.5, and **Shortcut after import** to **Genotyping**. The bar at the top counts the changes: **3 settings changed**.
- **Check the result.** **Only differing from parent** reduces the table to your changes, each beside its **Parent value**. **Reset to parent value** puts a parent's value back.

**Create** saves the set.

![Duplicate settings set, with 3 settings changed, Cancel and Create at the top. The Details card starts from Clarida defaults v2, with the name Genotyping 384-well and a description. In the Settings card, Only differing from parent is switched on, and the table lists three rows, each with a reset icon: Plate format, Design, All applications, 384-well where the parent value is 96-well; ΔCq threshold, QC, Genotyping and RQ, 0.3 where the parent has 0.5; and Shortcut after import, Workflow & display, All applications, Genotyping where the parent has Exploration](https://clarida.bio/images/methods/settings-sets-duplicate-v2.png)

Figure 2: The lab's set, derived from Clarida defaults, filtered to what differs from its parent.

See [Derived from a parent](https://clarida.bio/methods/settings-sets#derived-from-a-parent) and [What a set does not hold](https://clarida.bio/methods/settings-sets#what-a-set-does-not-hold).

### 3. Start experiments from it

A settings set reaches new experiments through a preset. In **Setup → Presets**, open your genotyping preset, and in its **Settings and layout** card choose **Replace settings set** and pick **Genotyping 384-well** (see [Presets](https://clarida.bio/methods/presets)). Every experiment started from that preset begins with 384-well plates and a ΔCq threshold of 0.3, without anyone entering either.

See [How an experiment gets its settings](https://clarida.bio/methods/settings-sets#how-an-experiment-gets-its-settings).

### 4. Check what an experiment started from

Open **Define**. Under **Active settings and layout**, the settings card names the set and version the experiment uses now, and counts the settings changed in this experiment, such as **3 overrides**. Once its set has been changed, a second line says from which set and when, such as **Switched from Lab QC v2 on Nov 2, 2026**. Its **View … in Experiment settings** shows the set's details.

![The setup area of Define. On the left, under Launched from, the Genotyping preset with a Clarida pill. On the right, under Active settings and layout, the settings card for Clarida defaults, with a Clarida pill, a v1 badge, a v2 available pill and the line 1 override · 192 local values, an eye icon and a change-set icon; below it the workflow layout card for Genotyping](https://clarida.bio/images/methods/layouts-define-card-v2.png)

Figure 3: Define names the experiment’s settings set, here Clarida defaults v1 with one override, above its workflow layout.

A setting changed inside the experiment, such as the ΔCq threshold in replicate QC, changes that experiment only. The setting then carries a marker that says who changed it and what the set holds; **Reset to set value** puts the set's value back.

See [The experiment records where it started](https://clarida.bio/methods/settings-sets#the-experiment-records-where-it-started) and [Changing a setting in one experiment](https://clarida.bio/methods/settings-sets#changing-a-setting-in-one-experiment).

### 5. Change a set that is in use

Six months on, the lab finds that 0.3 cycles flags too many replicate sets on its low-input samples, and settles on 0.4. Its set is used by its preset and a dozen experiments, so its values are locked. **Edit settings set** says why, and offers two ways on:

- **New version** continues the same set. It opens a draft of the next version; change the ΔCq threshold and choose **Save new version**. Nothing is saved before that.
- **Duplicate** makes a separate set, with its own name and versions.

The lab saves **v2**. Its preset now shows **v2 available** on the settings card. The preset has started experiments, so it is locked too: duplicate the preset, and in the copy click **v2 available** to move to v2. The dozen experiments keep 0.3: they started from v1, and v1 has not changed. To move one of them to 0.4, open it and choose **Update to v2** in its settings set's details.

See [Locked once used](https://clarida.bio/methods/settings-sets#locked-once-used).

### 6. Review what the parent changed

When Clarida publishes a new version of Clarida defaults, your set carries a **Parent updated** badge, and its view says so. Say a future v3 changes three settings the lab's set still inherits: the view reads **Clarida defaults v3 changed 3 settings this set still inherits · Review**. **Review** lists those settings with the old and the new parent value. Select the changes to adopt and choose **Complete review**; the rest are declined.

See [When the parent gets a new version](https://clarida.bio/methods/settings-sets#when-the-parent-gets-a-new-version).

## How it works

### What a settings set is

A settings set is a complete list of experiment settings, each with a value. It describes the experiment as a whole: thresholds, methods and modes that make sense for any experiment, whatever samples and assays it holds. Three properties follow from that:

- **A set is complete.** Every setting has a value. Where a setting may deliberately be empty, the empty value has a stated meaning, such as **No Cq floor** or **Auto per target**, and the table shows that meaning.
- **A set is copied, never linked.** When an experiment starts, the set's values are copied into it. Editing a set, or making a new version of it, never changes an experiment that already exists.
- **A set is versioned.** Once something uses a set, its values are locked. A change becomes a new version, and the old version stays exactly as it was, so what used it stays traceable to it.

### What a set holds

A set holds 75 settings, sorted into four groups:

| Group | What it covers | Examples |
| --- | --- | --- |
| **Design** | How the experiment is set up on the bench | PCR replicates, Plate format, Pipetting approach, Run layout mode and its scheme |
| **Analysis** | How measurements become results | Cq method, Baseline method, Cq aggregation, Normalization mode, Amplification efficiency mode, Scaling target |
| **QC** | When a result is flagged | Replicate QC metric, ΔCq threshold, Reference gene geNorm M threshold, Reference gene CV threshold, Negative control Cq floor |
| **Workflow & display** | How the workflow behaves and how results are shown by default | Exclusion note mode, Shortcut after import, Show genotypes as |

Each setting also names the **applications** it affects: RQ (relative quantification), Detection, Genotyping, Cq calling, or all of them. Plate format matters to every experiment; a reference gene threshold only to relative quantification; the end-point method to Cq calling and genotyping. The **Application** column's filter narrows the table to the settings your kind of experiment reads.

### Conditional settings

Some settings only matter when another setting selects them. The run layout scheme (Block by, Fill order, Replicate placement in sets and the rest) is used when the **Run layout mode** is **Assisted**. The **Normalization method** is used when the **Normalization mode** is **Manual**.

The **Conditional** column names the setting a row depends on, and its tooltip states the condition, for example **Applies when Normalization mode is not Auto**. Selecting the name takes you to that setting. Rows whose condition is not met right now are dimmed. They stay editable, since a set may hold a value for a mode it does not use, but the value has no effect until the mode selects it. **Only settings that apply now** hides them, which leaves the settings that take effect with the set's current choices.

### What a set does not hold

A set has to work in any experiment, so it never names a particular sample, target or well. Values that only make sense with one experiment's data stay in that experiment:

- which targets are your reference genes;
- a target's own efficiency, its own detection cut-off, or a per-well threshold;
- scaling to one named sample.

Where a set needs to point at items, it does so by role. **Scaling target** can be **Scale to calibrator**, the samples marked Calibrator, or **Scale to positive controls**; the experiment's samples supply who those are. **Amplification efficiency mode** set to **Assisted** with **Library first** looks for each target's efficiency in the Library before anywhere else.

This keeps a set portable: the same set serves a two-gene pilot and a ninety-sample study.

### Clarida defaults

Clarida's own recommendations are a settings set like any other, **Clarida defaults**, published by Clarida. It is the set of Clarida's own presets, including the one **Start a blank experiment** uses, and the usual parent of a lab's own set.

You can open it and inspect it setting by setting, with the same table, search and explanations as your own sets, but you cannot edit it. When Clarida changes a recommendation, the change arrives as a new version of Clarida defaults; the version your experiments started from stays as it was.

### Derived from a parent

A new set always starts from a parent: Clarida defaults, another set of your own, or a vendor's set whose publisher allows it. Every setting starts at the parent's value, so you change only what your set should do differently.

Whether a setting differs from its parent is not stored; it is computed by comparing the two sets. If you type the parent's value back in by hand, the row stops counting as a change. With **Only differing from parent** switched on, the table is a complete description of how your set differs from its parent. The set's **History** tab lists the same differences, each from the parent's value to yours.

### Locked once used

While nothing depends on it, a set you own can be edited freely. Its values lock as soon as:

- a preset or an experiment uses it;
- another set was derived from it;
- a newer version of it exists.

The editor says which applies, for example that the set is used by presets or experiments (deleted ones included), so its values are locked and what uses it stays traceable to it. Its name and description stay editable.

The reason is traceability. An experiment records the set and version it started from. If that version could change afterwards, the record would point at values the experiment never had. To change a locked set, you choose one of two actions, and Clarida never chooses for you:

- **New version** continues the same set. Presets that use it are offered the new version and do not move by themselves. An unused preset moves with one click on **v2 available**; a used preset is locked in turn, so you duplicate it (see [Presets](https://clarida.bio/methods/presets)).
- **Duplicate** makes a separate set with its own name and versions. Nothing else changes.

A set that nothing uses, with no newer version and no set derived from it, can be deleted. Otherwise it can be archived: it disappears from the Library and from the set choosers, while the presets and experiments that use it keep working.

### When the parent gets a new version

A derived set remembers which version of its parent it was last reconciled with. When the parent gets a newer version, the set carries **Parent updated**, its **History** tab says the parent has a newer version, and its view shows how many settings the parent's new version changed that the set still inherits, with **Review**.

Only settings you left at the parent's value are counted. Settings you changed are yours, and a parent's change to them is never raised. For the lab, Plate format, ΔCq threshold and Shortcut after import are never listed, whatever Clarida's new version does with them.

**Review** lists each of those settings with the value of the version you reconciled with and the value of the new version. Completing the review moves the set's reference point to the parent's new version, so a declined change is not raised again, and the settings you adopted no longer count as differing from the parent. If the set is unused, the adopted values are saved in place. If it is locked, they are saved as a new version, for the reasons above.

Nothing changes without that review. A parent's new version never alters a set derived from it on its own.

### Comparing two sets

Any two sets can be compared, not only a set and its parent. In a set's view, **Compare with…** places another set beside it as **Set B** and opens on **Only settings that differ**. Switching that off shows every setting side by side. **Stop comparing** returns to the set alone.

### How an experiment gets its settings

Every new experiment starts from a [preset](https://clarida.bio/methods/presets), and every preset references exactly one settings set. When the experiment is created, Clarida copies that set's values into it and records which set and version they came from. **Start a blank experiment** uses Clarida's default preset, which carries Clarida defaults.

A preset holds a reference to the set, not a copy. When the set gets a new version, the preset shows that a newer version is available, and its owner decides when to move. A new version never moves an experiment by itself: each keeps its values until someone changes its set in Define (see [Changing a setting in one experiment](https://clarida.bio/methods/settings-sets#changing-a-setting-in-one-experiment)).

### The experiment records where it started

Each experiment records the set and version it started from. **Define** shows the set it uses now under **Active settings and layout**, and after a change, which set it came from and when; the set's own details list, in turn, the presets and experiments that use it. The record travels with the results: the ELN export carries a **Started from** block in its overview, with the set, its version, and the set it was derived from.

The wording is deliberate. It says which set the experiment **started from**, not that it was analyzed with that set unchanged: its settings can still be changed inside the experiment (see the next section).

Experiments created before settings sets existed, experiments imported from qbase+ or RDML, and demo copies that could not keep their set are **attributed** to a set instead: the card says where they came from, such as **Imported from qbase+; compared with this set** or **Created before settings sets; compared with this set**, and never claims the experiment started from it. They keep the settings they came with. Where no set is recorded at all, the card says **No settings set recorded**.

### Changing a setting in one experiment

The experiment owns its copy of the settings. You can change any of them where the setting is used, for example the ΔCq threshold in replicate QC, and the change applies to this experiment alone. The set in the Library is untouched, and so is every other experiment started from it.

The ELN export keeps naming the set the experiment started from, and Define marks each changed setting. Where a value differs between the set and the experiment, the experiment's value is the one that computed the results.

To analyze with another set, use **Change settings set** on Define's settings card and pick the set. **Keep overrides (soft)** keeps the settings changed in this experiment and lists each one as the same as the new set or different from it; **Drop overrides (hard)** gives every setting the new set's value, cannot be undone, and says whose changes it drops. Either way the results are recomputed. When the set has a newer version, **Update to v2** in the set's details opens the same dialog on it and lists what the new version changes.

### Settings sets, layouts and presets

Three Library items decide how a new experiment starts, and each answers a different question:

- A **[workflow layout](https://clarida.bio/methods/workflow-layouts)** decides **which sections show, and where**.
- A **settings set** decides **how those sections compute**: the thresholds, methods and modes behind the numbers.
- A **[preset](https://clarida.bio/methods/presets)** brings them together, with any samples, assays, mixes or protocols the experiment should start with.

Layout and settings are independent. A detection layout and a genotyping layout can share one settings set, and hiding a section does not remove its settings: the experiment still receives the whole set.

## About the author

- Jan Hellemans

  Co-founder, Clarida

  Leads the science and the methods at Clarida. Published the qBase method in 2007 and co-authored the MIQE guidelines.

  [Google Scholar](https://scholar.google.com/citations?user=L8EeDgwAAAAJ)

## Every new experiment can start from your lab's own thresholds.

Duplicate Clarida defaults under Setup → Experiment settings; a preset carries it from there.

[Start free](https://app.clarida.bio/login) · [Talk to a founder](https://clarida.bio/talk-to-us?topic=help)
