db-alphafold

Retrieve AlphaFold protein structures by UniProt ID and download PDB/mmCIF files.

Updated Mar 13, 2026
One-click install
npx skills add https://github.com/biomaps-infra/blender-opencode --skill db-alphafold
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: db-alphafold
Source: https://github.com/biomaps-infra/blender-opencode/tree/main/.opencode/skills/db-alphafold
Command: npx skills add https://github.com/biomaps-infra/blender-opencode --skill db-alphafold

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires biopython, requests, numpy, matplotlib, pandas, scipy, google-cloud-bigquery, gsutil, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides access to a vast database of AI-predicted 3D protein structures, enabling researchers to analyze protein function and design new therapeutics even when experimental data is unavailable.

Core Features & Use Cases

  • Structure Retrieval: Get predicted structures for millions of proteins by UniProt ID.
  • Confidence Analysis: Evaluate the reliability of predictions using pLDDT and PAE metrics.
  • Bulk Data Access: Download entire proteomes or specific datasets via Google Cloud.
  • Use Case: A drug discovery team needs to analyze the structure of a novel kinase. They use this Skill to download the AlphaFold prediction, examine its active site confidence, and plan their inhibitor design strategy.

Quick Start

Use the db-alphafold skill to retrieve the structure for UniProt ID P00520.

Frequently Asked Questions about db-alphafold

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

FAQPage Schema
How do I retrieve a predicted protein structure using a UniProt ID?

To retrieve a predicted protein structure, query the AlphaFold database by providing the specific UniProt ID to download the corresponding 3D model in PDB or mmCIF format. This enables programmatic access for structural biology workflows without requiring experimental data.

What do pLDDT and PAE metrics tell me about AlphaFold structure predictions?

pLDDT and PAE metrics evaluate the reliability of AlphaFold structure predictions. The per-residue pLDDT score indicates local confidence, while the inter-residue PAE metric measures the predicted positional error between different protein domains, helping you assess active site accuracy.

Can I download entire proteomes from the AlphaFold database via Google Cloud?

Yes, you can download entire proteomes from the AlphaFold database via Google Cloud. This skill supports bulk data access using google-cloud-bigquery and gsutil, allowing you to retrieve large datasets for drug discovery and protein engineering workflows.

Does this AlphaFold database skill require Biopython to analyze 3D models?

Yes, this AlphaFold database skill requires Biopython along with numpy and pandas. These dependencies support the programmatic analysis of downloaded 3D protein structures and the evaluation of per-residue confidence metrics within your structural biology workflows.

What is the best way to access AI-predicted protein structures for drug discovery?

The best way to access AI-predicted protein structures for drug discovery is to programmatically query the AlphaFold database by UniProt ID. You can retrieve 3D models, analyze active site confidence metrics like pLDDT, and plan inhibitor design strategies.

When should I not rely on AlphaFold predictions for protein engineering?

You should carefully evaluate AlphaFold predictions using pLDDT and PAE confidence metrics before relying on them for protein engineering. Low confidence scores may indicate unreliable structural regions, requiring you to interpret the AI-predicted 3D models with caution during analysis.