Pdb Database
Download RCSB PDB coordinate files (mmCIF or PDB) for structural biology workflows and prototyping.
Overview
PDB Database is an agent skill for the Idea phase that downloads RCSB PDB coordinate files with sanitized IDs and rate-limited HTTP fetching.
Install
npx skills add https://github.com/google-deepmind/science-skills --skill pdb-databaseWhat is this skill?
- Downloads coordinate files from RCSB files-beta endpoint via shared HTTP client
- Sanitizes PDB IDs (including pdb_0000 prefix form) before fetch
- Supports comma-separated batch IDs with a 1,000-ID guardrail for bulk
- Python 3.10+ script with scienceskillscommon dependency
- Warns when bulk download should use official RCSB archives instead
- 1,000-ID ceiling before recommending official bulk archives
- HTTP client configured at 5.0 QPS toward RCSB files-beta
Adoption & trust: 572 installs on skills.sh; 1.7k GitHub stars; 3/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have PDB IDs but no quick, correct way to pull official coordinate files from RCSB into your local research environment.
Who is it for?
Indie researchers and builders prototyping structure-aware tools who need a few RCSB files via Python.
Skip if: Enterprise bulk PDB mirrors, workflows forbidden from external structure downloads, or teams without Python 3.10+.
When should I use this skill?
You need to download one or more RCSB PDB coordinate files by ID for research or prototype integration.
What do I get? / Deliverables
You obtain sanitized, API-fetched structure files for listed IDs while avoiding naive bulk downloads beyond the script’s guardrail.
- Downloaded mmCIF or PDB coordinate files for specified IDs
- Sanitized PDB ID list ready for downstream analysis
Recommended Skills
Journey fit
Fetching reference structures is early research work before you commit to a full structural analysis or ML pipeline. RCSB file retrieval supports literature and structure discovery in the research subphase rather than shipping or growth.
How it compares
Focused RCSB download script skill—not a full molecular dynamics suite or interactive structure viewer MCP.
Common Questions / FAQ
Who is pdb-database for?
Solo builders and researchers in structural biology or bioinformatics who need RCSB coordinate files during early exploration.
When should I use pdb-database?
In Idea research when validating a structure-based concept and you have specific PDB IDs to fetch before deeper analysis or model work.
Is pdb-database safe to install?
Review the Security Audits panel on this Prism page; the skill performs network fetches to RCSB—confirm licenses and data handling for your project.
SKILL.md
READMESKILL.md - Pdb Database
# Copyright 2026 Google LLC # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """Downloads PDB coordinate files (mmCIF or PDB) from RCSB.""" # /// script # requires-python = ">=3.10" # dependencies = [ # "scienceskillscommon", # ] # [tool.uv.sources] # scienceskillscommon = { path = "../../scienceskillscommon" } # /// import argparse import os import sys from science_skills.scienceskillscommon import http_client CLIENT = http_client.HttpClient("https://files-beta.rcsb.org", qps=5.0) def sanitize_id(pdb_id: str) -> str: """Sanitize PDB ID.""" pdb_id = pdb_id.lower().strip() if len(pdb_id) == 4: return f"pdb_0000{pdb_id}" return pdb_id def download_files(args: argparse.Namespace): """Downloads coordinate files.""" ids = [id.strip() for id in args.ids.split(",") if id.strip()] if len(ids) > 1_000: print( "Aborting: this script is not recommended for bulk downloads." "**Check with the user**, then consider downloading a copy of all" f" {args.format} files using the recommended bulk script at " "https://cdn.rcsb.org/wwpdb/docs/BetaArchiveBatchDownloader.py.", file=sys.stderr, ) return # RCSB bulk download script suggests a higher QPS may be acceptable here (~5). estimated_time_secs = len(ids) / 5.0 print(f"Estimated download time: {estimated_time_secs / 60:.1f} minutes).") if not os.path.exists(args.output_dir): os.makedirs(args.output_dir, exist_ok=True) for pdb_id in ids: sanitized_id = sanitize_id(pdb_id) shard_chars = sanitized_id[-3:-1] ext = "cif" if args.format == "mmcif" else "pdb" url = ( f"/pub/wwpdb/pdb/data/entries/{shard_chars}/{sanitized_id}/" f"structures/{sanitized_id}.{ext}.gz" ) output_path = os.path.join(args.output_dir, f"{sanitized_id}.{ext}.gz") print(f"Downloading {sanitized_id} from {url}...", file=sys.stderr) try: content = CLIENT.fetch_bytes(url) with open(output_path, "wb") as f: f.write(content) print(f"Saved to {output_path}", file=sys.stderr) except Exception as e: print(f"Failed to download {pdb_id}: {e}", file=sys.stderr) def parse_args() -> argparse.Namespace: """Parse command line arguments.""" parser = argparse.ArgumentParser(description="Download PDB coordinate files") parser.add_argument( "--format", type=str, required=True, choices=["mmcif", "pdb"], help="File format to download (mmcif or pdb)", ) parser.add_argument( "--ids", type=str, required=True, help="Comma-separated list of PDB IDs", ) parser.add_argument( "--output_dir", type=str, required=True, help="Directory to save files to", ) return parser.parse_args() if __name__ == "__main__": main_args = parse_args() download_files(main_args) # Copyright 2026 Google LLC # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """Gets PDB data using the RCSB Data API (GraphQL).""" # /// sc