Jean — Molecular Biology & Genomics

quantify_dpcr

Read-only

Digital PCR (dPCR) partition counts to absolute concentration, with no standard curve.

Input Schema

Name	Required	Description	Default
`wells`	Yes	the well table: `Well,Sample,Target,Positives,Partitions`. Instrument column names are accepted as they are printed (Bio-Rad's `AcceptedDroplets`, QIAcuity's `Valid partitions`), so a real export needs no renaming.
`_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.
`_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.
`ci_method`	No	which binomial confidence interval is carried onto the lambda scale. Values: agresti-coull (Agresti–Coull), asymptotic (Asymptotic (normal approximation)), bayes (Bayes (Jeffreys prior, highest-density)), cloglog (Complementary log-log), exact (Exact (Clopper–Pearson)), logit (Logit), lrt (Likelihood ratio), probit (Probit), prop.test (Score test inverted, continuity-corrected), profile (Profile likelihood), wilson (Wilson score). Advanced.	wilson
`volume_cv`	No	coefficient of variation of the partition volume, as a percentage. Ignored under the `poisson` model. There is no universal value and inventing one would fabricate the exact quantity the correction consumes - the one figure this repo has a published source for is QuantStudio 3D chip (Majumdar 2017) at 3.03 %, and that paper says the vendor re-estimates it per chip rather than treating it as a platform constant. In %. Only applies when volume_model is normal or truncated.
`conf_level`	No	confidence level for the interval. Advanced.
`ratio_target`	No	the assay whose COPY NUMBER you want, named as it appears in the table's Target column. Leave it and `ratio_reference` unset and no copy number is reported at all — a guessed pair is a fabricated comparison.
`volume_model`	No	how partition-volume SPREAD is handled (QP-5c). `poisson` is the standard estimator and assumes every partition holds the same volume. The two Poisson-Plus arms (Majumdar 2017) integrate the Poisson negative probability over a normal volume distribution and invert it: `normal` is the closed-form Eq (9), `truncated` the exact Eq (13) over a normal cut off at zero, which is the honest density because a partition cannot hold a negative volume. The spread biases the standard estimate DOWNWARD, hard at high occupancy and not at all at low. Values: poisson (Standard Poisson (no correction)), normal (Poisson-Plus, normal (Eq 9)), truncated (Poisson-Plus, truncated normal (Eq 13)).	poisson
`dilution_factor`	No	fold dilution applied to the sample before the reaction (2 = diluted 1:2). Advanced.
`ratio_reference`	No	the assay it is measured against: the single-copy or known-ploidy reference, again by its Target name.
`merge_replicates`	No	pool wells that share a sample and target before quantifying. Replicates of one measurement are one measurement, and pooling the COUNTS is the correct way to combine them, never averaging concentrations.
`reference_copies`	No	copies of the REFERENCE target per genome. 2 is a diploid autosome, 1 a haploid or single-copy locus. Advanced.
`reaction_volume_ul`	No	total reaction volume, for copies per reaction. In uL.
`template_volume_ul`	No	volume of template added to the reaction. Left at 0 the sample-concentration row is omitted rather than guessed. In uL. Advanced.
`partition_volume_nl`	No	mean volume of ONE partition. A per-platform constant, not a universal one (Bio-Rad QX200 approx 0.85 nL, QIAcuity 26k approx 0.34 nL, QuantStudio 3D approx 0.755 nL), and dMIQE 2020 requires it to be reported with the result. In nL.

Output Schema

Name	Required	Description
`rows`	Yes	one reading per well, or per pooled replicate group: occupancy, lambda with its interval, and every concentration the supplied volumes support
`count`	Yes	how many readings were produced
`ratios`	Yes	target-vs-reference copy number per sample, with the exact Poisson rate ratio and its interval rescaled onto the concentration scale the number is on. Empty when no assay pair was named — never a guessed pair
`flagged`	Yes	readings carrying at least one dMIQE occupancy flag
`ciMethod`	Yes	the interval method that produced the bounds: a CI without its method is not comparable
`confLevel`	Yes	the confidence level every interval in `rows` was computed at
`volumeModel`	Yes	how partition-volume spread was handled. `poisson` means it was not — a concentration without this is not comparable with one that was corrected
`volumeCvPercent`	Yes	the volume spread the correction used, echoed for the same reason as the partition volume: it changed the number
`partitionVolumeNl`	Yes	the partition volume the concentrations were computed at, echoed because dMIQE requires it to travel with the number

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedOct 21, 2026

addedOutput schema / properties / confLevel

Added value: +{
+  "description": "the confidence level every interval in `rows` was computed at",
+  "type": [
+    "number",
+    "null"
+  ]
+}

changedOutput schema / required

Previous value: -[
-  "rows",
-  "count",
-  "flagged",
-  "ciMethod",
-  "partitionVolumeNl",
-  "ratios",
-  "volumeModel",
-  "volumeCvPercent"
-]New value: +[
+  "rows",
+  "count",
+  "flagged",
+  "ciMethod",
+  "partitionVolumeNl",
+  "ratios",
+  "volumeModel",
+  "volumeCvPercent",
+  "confLevel"
+]

First observedOct 20, 2026

TDQS

B3.2/5.0

Behavior2/5

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 fully covered elsewhere. The description adds only the methodological note 'no standard curve' and says nothing about what the tool emits, when rows are omitted (the template_volume_ul/ratio_target omissions are documented only in the schema), or how replicate pooling changes the result.

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 clause with zero filler. It is efficient, though being a fragment rather than a structured statement it gives no scaffolding for the 15-parameter surface that follows.

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?

An output schema exists and the input schema is exhaustively annotated, so much is covered outside the description. Still, for a 15-parameter statistical tool with a nested _subject/_files object and an advanced CI/volume-model surface, one sentence is thin — it never mentions the well-table input shape or that copy-number ratios are an optional output.

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% and the per-property text is unusually rich (platform partition volumes, Poisson-Plus rationale, why guessed ratio pairs are fabricated), so the schema carries the parameter burden entirely. The description adds no parameter-level meaning of its own; baseline 3 applies.

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?

The one-line description gives a concrete operation and target: dPCR partition counts converted to absolute concentration, qualified by 'no standard curve', which effectively separates it from the qPCR-standard-curve siblings (call_cq, bootstrap_cq). It does not name any sibling tool outright, so it stops short of a 5.

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?

Usage is only implied: an agent must infer that this is the tool for digitisable partition-count data (Positives/Partitions wells). There is no statement of when-not to use it, no alternative for relative quantification (relative_expression) or for raw droplet classification (classify_droplets/gate_droplets), and no prerequisites mentioned.

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?