pdb-database

Search RCSB for 3D biomolecular structures and metadata via GraphQL.

2.7k|283|Updated May 13, 2026
One-click install
npx skills add https://github.com/google-deepmind/science-skills --skill pdb-database-google-deepmind
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pdb-database
Source: https://github.com/google-deepmind/science-skills/tree/main/skills/pdb_database
Command: npx skills add https://github.com/google-deepmind/science-skills --skill pdb-database-google-deepmind

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill helps you find and retrieve experimentally determined 3D biomolecular structures (proteins, nucleic acids, and bound ligands) together with useful metadata, avoiding manual navigation of RCSB entries.

Core Features & Use Cases

  • Attribute-based search: Discover the right RCSB attribute names via schema fetching, then run JSON searches for proteins, assemblies, polymers, chemicals, and experimental metadata.
  • Similarity-based search: Perform sequence, structure, motif, and chemical descriptor searches without requiring a schema fetch.
  • Metadata-only queries: Use GraphQL to fetch titles, experimental methods, taxonomy, sequence accessions, and other structured fields efficiently without downloading coordinate files.
  • Coordinate file download: Download mmCIF/PDB coordinate files for specific entry IDs using the provided downloader script, with guardrails against bulk downloads.
  • Query transparency: Produces output guidance that explains what the query did so users can correct assumptions.

Quick Start

Use the pdb-database skill to search RCSB for entries that match a structure query and write the JSON results to a file named results.json.

Frequently Asked Questions about pdb-database

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

FAQPage Schema
How do I search RCSB for biomolecular structures by sequence or structural similarity?

Metadata-only GraphQL queries allow you to fetch specific fields like experimental methods, resolution, and taxonomy directly. This avoids downloading full mmCIF coordinate files when atomic data is unnecessary.

What is the best way to download mmCIF coordinate files for specific RCSB entries?

The best way to download mmCIF coordinate files is by using the provided rate-limited downloader script. This method retrieves atomic data for specific entry IDs while applying guardrails against bulk downloads.

Can I perform attribute-based searches for experimental metadata in RCSB without knowing exact field names?

Yes, you can fetch the RCSB schema to discover valid searchable attribute names. Once identified, you run structured JSON searches for proteins, assemblies, polymers, and experimental metadata.

Are there limitations on bulk downloading PDB coordinate files?

Yes, strict rate-limiting guardrails are applied to prevent bulk downloading of coordinate files. The provided scripts are designed for targeted retrieval of specific entry IDs rather than large-scale data extraction.