Skip to main content
Glama
smaniches

AlphaFold Sovereign MCP

by smaniches
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

Variant 3-D Structural Triage

triage_variant_3d
Read-onlyIdempotent

Triage an HGVS variant by assembling source-reported ClinVar pathogenicity, gnomAD gene-constraint scores, and context notes into one prioritised evidence report.

Instructions

Retrieve source-reported variant evidence for research.

Fuses the upstream signals this tool currently wires into a single prioritised report:

  1. Pathogenicity — ClinVar interpretation + review status. The alphamissense_score / alphamissense_interpretation fields are always null / "Not available" here: AlphaMissense is not wired into this tool. For an AlphaMissense pathogenicity score use generate_variant_clinical_report.

  2. Population genetics — gnomAD LOEUF / pLI gene-constraint scores. Per-variant allele frequencies and the per-ancestry breakdown are not wired into this tool.

  3. Disease associations — a placeholder note pointing at get_target_diseases(); the Open Targets / MONDO traversal is a roadmap (Wave-3) item.

  4. Structural context — a text note pointing at analyze_structural_confidence (resolve the gene to a UniProt accession first); the AlphaFold pLDDT / PAE join into this report is a roadmap (Wave-3) item.

Returns the legacy pathogenicity_tier key as NOT_ASSESSED. Raw verified ClinVar classifications remain source-attributed in the pathogenicity section; they are not converted into a local risk tier. Gene-relative candidates remain explicitly unverified. AlphaMissense is not wired into this particular tool. The report is not a calibrated risk estimate.

Example: triage_variant_3d(hgvs='BRCA1:c.181T>G')

Input Schema

NameRequiredDescriptionDefault
paramsYes

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv1.5.9
    • changedInput schema / properties / params / properties / hgvs / description
      Previous value: -"HGVS variant expression. Supports coding DNA notation e.g. 'BRCA1:c.181T>G', 'TP53:c.817C>T', or genomic e.g. 'chr17:g.43094692G>A'."New value: +"HGVS variant expression, e.g. 'BRCA1:c.181T>G' for candidate discovery or 'NM_007294.4(BRCA1):c.181T>G' for versioned-transcript identity matching. Gene-only coordinates cannot verify allele identity."
  2. Changed3 schema fields changedv1.2.0
    • changedInput schema / properties / params / properties / include_disease_context / description
      Previous value: -"Fetch MONDO disease + Open Targets evidence."New value: +"Add a placeholder disease-context note. The MONDO / Open Targets traversal is a Wave-3 roadmap item (stub)."
    • changedInput schema / properties / params / properties / include_gnomad / description
      Previous value: -"Fetch population allele frequencies from gnomAD."New value: +"Fetch gnomAD gene-CONSTRAINT scores (LOEUF / pLI) only. Per-variant allele frequencies are not wired into this tool."
    • changedInput schema / properties / params / properties / include_structure / description
      Previous value: -"Retrieve AlphaFold structure for context."New value: +"Add a pointer note toward structural-confidence analysis. The full AlphaFold pLDDT/PAE join is a Wave-3 roadmap item (not wired into this tool yet)."
  3. First observedv1.1.8

TDQS

A4.2/5.0
Behavior5/5

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

Annotations cover safety (readOnly/idempotent/openWorld), and the description goes well beyond them: it discloses that alphamissense fields are always null, that pathogenicity_tier is returned as NOT_ASSESSED, that the structural and disease sections are stubs, and that gene-relative candidates stay unverified and the output is not a calibrated risk estimate. That is unusually candid disclosure of degraded/degrading behaviour.

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

Conciseness3/5

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

The purpose is front-loaded and the numbered structure is scannable, but the text is padded with repetition: AlphaMissense is disclaimed three separate times and 'Wave-3 roadmap' is repeated across items 3 and 4. Roughly a third of the prose restates earlier caveats.

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

Completeness5/5

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

An output schema exists, so return values need no elaboration; what the description must supply is the trust profile of the report, and it does: which sections are real, which are stubs, which fields are always null, and how to get the missing data via siblings. Nothing needed to call or interpret it correctly is absent.

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

Parameters4/5

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

Reported schema coverage is 0%, but the description compensates: it gives a worked HGVS example, clarifies that gene-only coordinates cannot verify allele identity, and explains what each flag actually yields (LOEUF/pLI only, a pointer note, a placeholder note). This meaningfully extends the schema, though it never discusses the undocumented 'params' wrapper.

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 opening sentence names a specific verb and resource ('Retrieve source-reported variant evidence') and the numbered list enumerates exactly which signals are fused into the report. It is clearly separable from generate_variant_clinical_report and analyze_structural_confidence, though the '3-D Structural' title oversells a tool whose structural component is only a pointer note.

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

Usage Guidelines4/5

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

It explicitly routes each missing capability to the correct sibling: AlphaMissense scoring to generate_variant_clinical_report, structural confidence to analyze_structural_confidence (with the note to resolve a UniProt accession first), and disease context to get_target_diseases(). It does not state the positive case for choosing this tool over query_variant_database or classify_variant_acmg.

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

Deploy Server

Other Tools