How do I find a protein structure if I don't have the UniProt ID?

You can use the built-in UniProt search capabilities to query by protein name or gene name to retrieve the correct accession ID before fetching the structure.

Can I download data for an entire species at once?

Yes, the skill provides patterns for bulk downloading complete proteomes via Google Cloud Storage using taxonomy IDs for high-throughput analysis.

How do I interpret the AlphaFold confidence scores?

AlphaFold provides pLDDT scores (0-100). Scores above 90 represent very high confidence, while scores below 50 suggest the region may be disordered or unstructured.

What file formats are supported for protein structures?

The skill supports downloading atomic coordinates in both mmCIF (PDBx) and PDB formats, as well as JSON files for confidence scores and error matrices.

AlphaFold Database Access

Name: AlphaFold Database Access
Author: Zehong-Wang

byZehong-Wang

0•

Data Science & ML

Retrieve and analyze over 200 million AI-predicted protein structures from the AlphaFold DB for structural biology and drug discovery.

This skill provides a specialized interface for interacting with the AlphaFold Database, a massive repository of AI-generated protein structures maintained by DeepMind and EMBL-EBI. It enables researchers and developers to programmatically fetch structure predictions using UniProt IDs, download coordinate files in mmCIF or PDB formats, and interpret critical confidence metrics like pLDDT and PAE. Whether you are building computational pipelines for drug discovery, performing large-scale proteome analysis via Google Cloud, or integrating structural data into bioinformatics workflows, this skill streamlines the acquisition and analysis of high-quality protein models.

Key Features

01Retrieve 3D protein structures by UniProt ID or protein name

02Download high-resolution PDB and mmCIF coordinate files

030 GitHub stars

04Analyze prediction confidence metrics including pLDDT and PAE matrices

05Access bulk proteome datasets through Google Cloud Platform integration

06Parse and process structural data using Biopython and NumPy

Use Cases

01Accelerating drug discovery by identifying druggable pockets in predicted protein structures

02Designing protein variants and mutations by analyzing high-confidence structural regions

03Conducting large-scale structural proteomics across entire species or taxonomies

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zehong-wang/kosmos alphafold-database

For use in Claude.ai and ChatGPT

Download Skill