Jean — Molecular Biology & Genomics

relative_expression

Read-only

Relative gene expression from a Ct table, by all THREE models the Analyze panel offers: delta-delta-Ct (normalised to a reference gene AND calibrated to a reference group), delta-Ct (calibrated to a reference group only), and the two-step standard-curve model, which fits each gene's own dilution series and converts Ct to amount before taking the ratio.

Input Schema

Name	Required	Description	Default
`tube`	No	`separate_tube`: target and reference amplified in different wells, errors independent. `same_tube`: multiplexed in one well, subtracted per well before averaging. Advanced.	separate_tube
`model`	No	which model. `delta_delta_ct` and `relative_curve` need a reference gene, `delta_ct` does not; `relative_curve` additionally needs `standard_ct_table` and `amounts`, because it fits a real dilution series rather than assuming efficiency. Values: delta_delta_ct (Delta-delta-Ct (normalised + calibrated)), delta_ct (Delta-Ct (calibrated only)), relative_curve (Standard curve (fitted amounts)).	delta_delta_ct
`_files`	No	The project: record references shaped like _subject. Their records become library documents a tool names in sequences/parts/document; a trace or alignment is found by file name.
`groups`	Yes	the group each sample belongs to, one per line, in the same order as the Ct rows (e.g. six lines of `control` then six of `treated`).
`amounts`	No	the input amount of each dilution row, one per row of `standard_ct_table`, whitespace- or comma-separated. The fit is against log10 of these, so every value must be positive. Advanced.
`lm_mode`	No	how `lm.design` normalises the target against the reference gene before fitting: `subtract` the reference Ct (the delta-Ct scale), or `divide` by it. Values: subtract (Subtract (ΔCt = target − reference)), divide (Divide (target ÷ reference)). Only applies when sig_test is lm.design. Advanced.	subtract
`_subject`	No	The record to act on: exactly one of sequence, accession (+source), sample (an id from list_samples) or file ({name, text} or {name, base64}); record_index picks one record of several.
`ct_table`	Yes	the Ct table: a header row of gene names, then one row per sample. Tab- or comma-separated, the same table the Analyze tab takes.
`sig_test`	No	which significance test runs beside the model. `t.test` is Welch's two-sample test on the normalised values; `lm` is the equivalent linear model, which reports the same p with a confidence interval; `wilcox.test` is the rank-sum test, for plates too small or too skewed to assume normality; `lm.design` fits `lm(norm ~ model_matrix + 0)` for designs the two-group form cannot express, and is reported through `design` rather than the test table. Values: none (None), t.test (Welch t-test), lm (Linear model (two-group)), wilcox.test (Wilcoxon rank-sum), lm.design (Design-matrix linear model). Advanced.	none
`lm_formula`	No	the design formula for `lm.design`, in R's own notation (`~ group`, `~ group + dose`, `~ group * run`). `group` is always available from `groups` with no paste at all. Only applies when sig_test is lm.design. Advanced.	~ group
`lm_variables`	No	extra design variables for `lm.design`: a header row of variable names, then one row per sample in the Ct table's order. A column whose every value parses as a number is a covariate, otherwise a factor - the same distinction R makes. Leave empty to model on `group` alone. Only applies when sig_test is lm.design. Advanced.
`wilcox_exact`	No	which Wilcoxon null distribution. `auto` applies R's own rule (exact while both groups are under 50 and there are no ties); `exact` forces the exact distribution, which is costly on large groups; `normal` forces the approximation. Values: auto (Auto (exact if both groups < 50)), exact (Exact conditional), normal (Normal approximation). Only applies when sig_test is wilcox.test. Advanced.	auto
`lm_conf_level`	No	confidence level for the `lm.design` coefficient intervals. Only applies when sig_test is lm.design. Advanced.
`reference_gene`	No	the gene expression is normalised to. Required for delta-delta-Ct and for the standard-curve model; ignored by delta-Ct, which normalises to a group only. Only applies when model is delta_delta_ct or relative_curve.
`wilcox_correct`	No	tail refinement for the Wilcoxon normal approximation: `none`, R's `continuity` correction, or 1-3 Edgeworth terms. Ignored when the exact distribution is used. Values: none (None), continuity (Continuity), 1 (Continuity + Edgeworth 1), 2 (Continuity + Edgeworth 2), 3 (Continuity + Edgeworth 3). Only applies when sig_test is wilcox.test. Advanced.	continuity
`reference_group`	Yes	the group everything is calibrated against: the untreated / time-zero / wild-type arm.
`assess_efficiency`	No	run `pcr_assess(method = 'efficiency')` beside delta-delta-Ct: delta-Ct against the reference gene regressed on log10 amount, per gene. This is how the model's equal-efficiency assumption is CHECKED rather than assumed - a slope far from zero says delta-delta-Ct is the wrong model for this plate. Needs `standard_ct_table` and `amounts`. Advanced.
`standard_ct_table`	No	the SERIAL-DILUTION Ct table the standard curve is fitted to - a second plate, not the sample one. Required by `relative_curve`, and by `assess_efficiency` on any model. Advanced.
`wilcox_conf_level`	No	confidence level for the Wilcoxon location-shift interval. Only applies when sig_test is wilcox.test. Advanced.
`wilcox_alternative`	No	the Wilcoxon alternative hypothesis. Values: two.sided (Two-sided), less (Less (shift < 0)), greater (Greater (shift > 0)). Only applies when sig_test is wilcox.test. Advanced.	two.sided

Output Schema

Name	Required	Description
`rows`	Yes	one row per group x gene: the normalised and calibrated delta-Ct (or fitted amount ratio), the fold change and its bounds
`test`	Yes	the significance test beside the model - one row per gene, with the statistic, p value and interval - or null when none was asked for
`count`	Yes	how many rows the model produced
`genes`	Yes	the genes measured, reference gene excluded under delta-delta-Ct
`model`	Yes	the model that produced them
`trend`	Yes	the per-gene standard curve (intercept, slope, R-squared) the `relative_curve` model consumed, or null
`design`	Yes	the design-matrix linear model: the formula, the columns it built, and per gene a coefficient table WITH the fit diagnostics. Null unless `sig_test` is `lm.design`
`groups`	Yes	the groups, in the order the rows use
`efficiency`	Yes	the equal-efficiency check: delta-Ct regressed on log10 amount per gene. A slope far from zero says delta-delta-Ct is the wrong model here. Null unless `assess_efficiency` ran
`referenceGene`	Yes	the gene every target was normalised to under delta-delta-Ct; null for the models that use none
`standardPoints`	Yes	the dilution observations behind that curve, with fitted values and residuals - a standard curve is judged by whether a dilution sits off the line, not by R-squared

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedOct 21, 2026

addedOutput schema / properties / referenceGene

Added value: +{
+  "description": "the gene every target was normalised to under delta-delta-Ct; null for the models that use none",
+  "type": [
+    "string",
+    "null"
+  ]
+}

changedOutput schema / required

Previous value: -[
-  "rows",
-  "count",
-  "model",
-  "genes",
-  "groups",
-  "test",
-  "design",
-  "trend",
-  "standardPoints",
-  "efficiency"
-]New value: +[
+  "rows",
+  "count",
+  "model",
+  "genes",
+  "groups",
+  "test",
+  "design",
+  "trend",
+  "standardPoints",
+  "efficiency",
+  "referenceGene"
+]

First observedOct 20, 2026

TDQS

B3.4/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds that the standard-curve model fits each gene's own dilution series and converts Ct to amount before ratioing, which is useful behavioural context, but says nothing about errors, what happens with insufficient replicates, or that a whole significance-testing surface runs alongside the model.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A single front-loaded sentence with no filler; the purpose comes first and the three model variants follow. It is dense but every clause maps to a real distinction among models. A slight readability cost from cramming three model definitions into one sentence without structure.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With an output schema and full schema descriptions, the description need not explain return values, and it adequately covers the core model choice. However, for a 20-parameter tool it omits the entire significance-testing half of the surface (t.test/lm/wilcox.test/lm.design, design matrices, confidence levels), so an agent reading only the description would not know the tool also performs statistical testing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% across all 20 parameters, so the schema already carries the parameter semantics and the baseline is 3. The description largely restates what the `model` enum description already explains about reference gene and standard-curve requirements, adding no format, ordering, or validation detail beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Relative gene expression from a Ct table') and scopes it to the three model families (delta-delta-Ct, delta-Ct, standard curve), which tells an agent exactly what computation is produced. It does not name a sibling tool (e.g. call_cq, rank_reference_genes, differential_expression) or say how this differs from them, so differentiation is left to inference.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description does explain the prerequisites that select a model implicitly ('normalised to a reference gene AND calibrated to a reference group' vs 'calibrated to a reference group only' vs fitting a dilution series), but it never states when to use this tool rather than a sibling, nor any exclusions. Real when-to-use guidance lives in the schema's `model` and `sig_test` descriptions, not in the top-level description.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Jean — Molecular Biology & Genomics tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?