skills.rest
mims-harvard
mims-harvardOFFICIAL
Explore all skills by this author →
Website𝕏@marinkazitnikGitHub

tooluniverse-protein-lof-mechanism

Synthesizes five computational signals to propose loss-of-function mechanisms for missense variants.

1.7k|254|Updated Mar 3, 2025
Search Tags:#alphafold#variant interpretation#alphamissense#loss-of-function#missense variant#protein stability#mechanism synthesis
One-click install
npx skills add https://github.com/mims-harvard/ToolUniverse --skill tooluniverse-protein-lof-mechanism
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tooluniverse-protein-lof-mechanism
Source: https://github.com/mims-harvard/ToolUniverse/tree/main/plugins/tooluniverse/skills/tooluniverse-protein-lof-mechanism
Command: npx skills add https://github.com/mims-harvard/ToolUniverse --skill tooluniverse-protein-lof-mechanism

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires esm.

What problem does it solve?

A pathogenicity score alone does not explain why a missense variant damages a protein. This Skill integrates five independent computational evidence layers to propose a specific molecular loss-of-function mechanism, distinguishing structural destabilization from direct functional disruption of catalytic, binding, PTM, or interface sites.

Core Features & Use Cases

  • Multi-signal evidence synthesis: Combines AlphaMissense pathogenicity, AlphaFold structural context (pLDDT), ESMC sequence likelihood, SAE feature disruption, and DynaMut2 stability (ΔΔG) into one mechanistic hypothesis.
  • Mechanism classification: Maps signal patterns to six LoF categories (structural stability, catalytic, binding, PTM, interface, generic damaging) with confidence grading and honest limitation reporting.
  • Use Case: Given TP53 R175H, the Skill resolves the UniProt accession, runs all five analyses, and reports that the variant destabilizes the fold (structural stability LoF) with high confidence, informing drug-rescue strategy decisions.

Quick Start

Analyze the loss-of-function mechanism of variant P04637 R175H by integrating AlphaMissense, AlphaFold, ESMC, SAE, and DynaMut2 evidence.

Frequently Asked Questions about tooluniverse-protein-lof-mechanism

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I determine the loss-of-function mechanism of a missense variant?▼

Provide a variant as accession plus ref/position/alt (e.g., P04637 R175H). The workflow runs AlphaMissense, AlphaFold pLDDT, ESMC sequence likelihood, SAE feature disruption, and DynaMut2 ΔΔG, then maps the combined signal pattern to one of six mechanism categories.

What is the difference between structural stability LoF and direct functional disruption?▼

Structural stability LoF means the mutation destabilizes the fold (ΔΔG above +1 kcal/mol with negative ΔlogP), causing misfolding or degradation. Direct functional disruption means the protein folds normally but a catalytic, ligand-binding, PTM, or interface site is broken, indicated by SAE feature loss with near-zero ΔΔG.

DynaMut2 vs ThermoMPNN for protein stability prediction?▼

DynaMut2 is the default because it is wired into ToolUniverse via a hosted academic API with no setup. ThermoMPNN offers higher accuracy near the ΔΔG threshold and handles double mutants and indels, but requires a local GPU install or a commercial hosted API key.

Does this workflow require an API key?▼

Only the SAE feature disruption signals require an ESM_API_KEY from EvolutionaryScale, plus the esm package with SAE support. AlphaMissense, AlphaFold, UniProt, and DynaMut2 are hosted endpoints that need no API key.

What are the limitations of SAE-based variant interpretation?▼

SAE feature labels are inferred from a 10-protein panel, not curated, so some features remain uncategorized. The analysis window is ±8 residues, so long-range allosteric effects are missed, and SAE outputs carry a non-commercial license.

When should I not use this missense LoF mechanism workflow?▼

Do not use it for non-coding or regulatory variants, ACMG pathogenicity classification, or cancer-specific driver analysis, which have dedicated skills. It also only supports single canonical isoforms and missense changes, not indels, nonsense, or splice variants.

Related Skills

Explore More Education & Research Skills→
MDheweiMDhewei
64

protein-structure-for-gene

Resolve gene symbols to UniProt accessions and render domain maps.

Community
Intermediate
monarch-initiativemonarch-initiative
50

cancer-curator

Curate cancer entries with genetic drivers, pathophysiology, and ontology bindings.

Official
Advanced
exon-researchexon-research
475

genomi-gnomad

Fetch population allele frequencies for a specified variant from gnomAD.

Official
Basic
nahisahonahisaho
3

scientific-variant-interpretation

Automate ACMG-aligned clinical interpretation of genetic variants and generate standardized reports.

Community
Advanced