skills.rest
K-Dense-AI
K-Dense-AIOFFICIAL
Explore all skills by this author →
Website𝕏@k_dense_aiGitHub

alphafold-database

Retrieve AlphaFold-predicted protein structures and confidence metrics by UniProt IDs.

33.0k|3.2k|Updated Oct 19, 2025
Search Tags:#protein structure#UniProt#pLDDT#AlphaFold#PDB#mmCIF#PAE#AI prediction
One-click install
npx skills add https://github.com/K-Dense-AI/claude-scientific-skills --skill alphafold-database
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: alphafold-database
Source: https://github.com/K-Dense-AI/claude-scientific-skills/tree/main/scientific-databases/alphafold-database
Command: npx skills add https://github.com/K-Dense-AI/claude-scientific-skills --skill alphafold-database

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Obtaining 3D protein structures for research is often limited by time-consuming experimental methods. This Skill provides automated access to AlphaFold DB's 200M+ AI-predicted structures, accelerating drug discovery, protein engineering, and structural biology studies by providing immediate structural insights.

Core Features & Use Cases

  • Structure Retrieval: Get protein structures by UniProt ID in PDB/mmCIF formats, including full atomic coordinates.
  • Confidence Analysis: Analyze per-residue pLDDT scores and Predicted Aligned Error (PAE) matrices to rigorously assess prediction reliability and domain arrangements.
  • Bulk Data Access: Download entire proteomes via Google Cloud Platform for large-scale structural genomics and comparative analysis.
  • Use Case: Automatically download the predicted structure for a target protein, extract its pLDDT scores to identify high-confidence regions, and visualize its PAE matrix to understand domain arrangements, all within minutes for rapid drug target assessment.

Quick Start

Get all predictions for a UniProt accession (e.g., P00520)

from Bio.PDB import alphafold_db predictions = list(alphafold_db.get_predictions("P00520")) print(f"Found {len(predictions)} predictions for P00520.")

Download the first structure file (mmCIF format)

if predictions: cif_file = alphafold_db.download_cif_for(predictions[0], directory="./structures") print(f"Downloaded: {cif_file}")

Frequently Asked Questions about alphafold-database

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

FAQPage Schema
How do I download protein structures from AlphaFold DB using UniProt IDs?▼

Retrieve AlphaFold-predicted protein structures by UniProt ID through REST API endpoints that return PDB or mmCIF coordinate files. Use the Skill to query by accession number, download atomic structures directly, and access metadata in minutes for immediate structural analysis.

What are pLDDT scores and PAE matrices, and why do they matter for protein predictions?▼

pLDDT scores measure per-residue confidence in AlphaFold predictions, while PAE matrices show predicted alignment error between residues. Together they assess prediction reliability and domain arrangements, helping you identify high-confidence regions and validate structural models before downstream analysis.

Can I download entire proteomes or bulk datasets for structural genomics studies?▼

Yes. This Skill accesses AlphaFold DB's 200M+ structures via Google Cloud Platform, enabling bulk downloads of complete proteomes for large-scale comparative analysis, drug discovery pipelines, and structure-based research without time-consuming experimental methods.

How do I integrate AlphaFold predictions into an automated drug discovery workflow?▼

Automate structure retrieval by UniProt ID, extract confidence metrics and PAE data programmatically, and feed results into computational pipelines. The Skill returns REST API endpoints and direct file URLs with metadata, enabling seamless integration into screening and target assessment workflows.

Does this work with existing Python bioinformatics tools like BioPython?▼

Yes. The Skill integrates with BioPython, numpy, pandas, scipy, and matplotlib for coordinate parsing, statistical analysis, and visualization. It also supports Google Cloud BigQuery and gsutil for cloud-scale data access and analysis.

What's the fastest way to assess prediction quality across multiple protein targets?▼

Retrieve structures for multiple UniProt IDs in parallel, extract pLDDT and PAE metrics directly from AlphaFold DB, and visualize confidence profiles using built-in analysis tools. This accelerates target prioritization and domain analysis for rapid drug discovery decisions.

Related Skills

Explore More Education & Research Skills→
AlterLab-IEUAlterLab-IEU
58

alterlab-alphafold-db

Retrieve AlphaFold predictions and model files by UniProt accession.

Community
Advanced
ovachieverovachiever
52

pdb-database

Search PDB structures and download coordinates and metadata via REST or GraphQL.

Community
Advanced
silversteinsilverstein
21

alphafold-database

Retrieve AlphaFold protein structures and confidence metrics from UniProt accessions.

Community
Intermediate
ClawBioClawBio
1.1k

struct-predictor

Predict protein structures locally and compare them to PDB references.

Official
Advanced