alphafold-database-access

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

298|27|Updated Feb 18, 2026
One-click install
npx skills add https://github.com/jaechang-hits/SciAgent-Skills --skill alphafold-database-access
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: alphafold-database-access
Source: https://github.com/jaechang-hits/SciAgent-Skills/tree/main/skills/structural-biology-drug-discovery/alphafold-database-access
Command: npx skills add https://github.com/jaechang-hits/SciAgent-Skills --skill alphafold-database-access

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill provides access to AlphaFold DB's vast collection of AI-predicted protein structures, enabling researchers to quickly retrieve and analyze 3D protein models for proteins lacking experimental data.

Core Features & Use Cases

  • Retrieve Structures: Get predicted structures by UniProt ID.
  • Download Files: Obtain PDB or mmCIF coordinate files.
  • Analyze Confidence: Evaluate prediction quality using pLDDT and PAE metrics.
  • Bulk Access: Download entire proteomes via Google Cloud.
  • Use Case: A drug discovery scientist needs to investigate potential binding sites on a protein that has no experimental structure. They use this Skill to download the high-confidence AlphaFold prediction and analyze its active site.

Quick Start

Use the alphafold-database-access skill to download the structure for UniProt ID P00520.

Frequently Asked Questions about alphafold-database-access

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

FAQPage Schema
How do I download AlphaFold predicted protein structures using a UniProt ID?

To download AlphaFold predicted protein structures, provide a UniProt ID to retrieve the corresponding PDB or mmCIF coordinate files. This allows direct access to AI-predicted 3D protein models for proteins lacking experimental data.

What is the best way to evaluate AlphaFold prediction confidence for a specific protein?

Evaluating AlphaFold prediction confidence involves analyzing per-residue pLDDT scores and visualizing inter-residue Predicted Aligned Error (PAE) matrices. These metrics help determine the reliability of specific structural regions and potential binding sites.

Can I use BioPython to directly parse AlphaFold structures retrieved by UniProt ID?

Yes, you can use BioPython to directly parse AlphaFold structures. The Skill integrates with BioPython for direct structure parsing, allowing immediate structural analysis of the retrieved PDB or mmCIF coordinate files.

Does this support bulk downloading of entire proteomes via Google Cloud Storage?

Yes, it supports bulk downloading of entire proteomes via Google Cloud Storage. This feature enables researchers to access large datasets of AI-predicted protein structures efficiently for proteome-wide structural biology studies.

How do I analyze active sites for drug discovery when experimental structures are unavailable?

To analyze active sites for drug discovery without experimental structures, retrieve the high-confidence AlphaFold prediction using the UniProt ID. You can then download the coordinate files and evaluate the active site using pLDDT confidence metrics.