PDBe Protein Structure Annotations Scraper

Scrape experimental protein structures from PDBe — the Protein Data Bank in Europe — into clean, analysis-ready JSON. Give it a list of PDB accession codes and get back the experimental method, resolution, R-factor, space group, deposition/release dates, biological assembly, polymer entities, bound ligands, source organism, primary publication, and SIFTS UniProt mappings — all merged into one record per structure. No API key. No login. ~$0.5 per 1,000 structures.

Built on the free PDBe REST API (EMBL-EBI, Hinxton, UK).

What you get per structure

Field	Description
pdb_id, title, url	Accession, title, PDBe entry link
experimental_methods	e.g. ["X-ray diffraction"], ["Electron Microscopy"], ["Solution NMR"]
resolution	Ångström, float — null for NMR/EM without a reported resolution (never faked as 0)
r_factor, space_group	Refinement quality / crystal symmetry (null for non-diffraction methods)
deposition_date, release_date, revision_date	ISO YYYY-MM-DD
entity_count, assembly_count, preferred_assembly	Entity totals + the biologically preferred assembly (name/form)
polymer_entities	Compact [{name, type, length, chains}] for polypeptide/nucleotide chains
ligands, ligand_count, has_ligands	Bound chemical components ([{id, name}])
uniprot_accessions, uniprot_count	SIFTS-mapped UniProt accessions
organism, title_publication, pubmed_id, doi	Source organism + primary citation
source, scraped_at	Provenance

Input

{
  "mode": "entries",
  "pdbIds": ["1cbs", "4hhb", "6vxx"],
  "includeLigands": true,
  "includeUniprot": true,
  "maxItems": 100
}

• mode — entries (PDB id list → merged records).
• pdbIds — 4-character PDB codes (case-insensitive).
• includeLigands / includeUniprot — toggle the extra ligand / SIFTS requests.
• maxItems — cap (default 100, max 2000).

Example output (6vxx — SARS-CoV-2 spike, cryo-EM)

{
  "type": "structure",
  "pdb_id": "6vxx",
  "title": "Structure of the SARS-CoV-2 spike glycoprotein (closed state)",
  "experimental_methods": ["Electron Microscopy"],
  "resolution": 2.8,
  "r_factor": null,
  "space_group": null,
  "release_date": "2020-03-11",
  "preferred_assembly": {"assembly_id": "1", "name": "trimer", "form": "homo"},
  "polymer_entities": [{"name": "Spike glycoprotein", "type": "polypeptide(L)", "length": 1288, "chains": ["A"]}],
  "ligands": [{"id": "NAG", "name": "N-ACETYL-D-GLUCOSAMINE"}],
  "uniprot_accessions": ["P0DTC2"],
  "organism": "Severe acute respiratory syndrome coronavirus 2",
  "url": "https://www.ebi.ac.uk/pdbe/entry/pdb/6vxx",
  "source": "pdbe"
}

Use cases

• Structural biology — bulk-pull method, resolution and assembly metadata across a set of structures.
• Drug discovery — enumerate bound ligands and their chem-comp ids for target/pocket analysis.
• Structure QC — compare resolution, R-factor and validation-relevant fields across candidates.
• Teaching — hand students clean, merged structure records without wrangling six endpoints.

How this differs from the RCSB and AlphaFold scrapers

The Protein Data Bank is a global archive with regional data-in / data-out partners. This actor deliberately targets the European (PDBe) aggregated views, which are not the same as the US RCSB representation:

• PDBe Structures Scraper (this actor) — uses the PDBe REST API. You get PDBe's aggregated per-entry views and, notably, SIFTS UniProt mappings (residue-level structure↔sequence cross-references maintained at EMBL-EBI), the "preferred" biological assembly, and PDBe's molecule/ligand aggregation. European mirror, European field shapes.
• RCSB PDB Scraper — uses the US RCSB data API, which exposes a different JSON model and different aggregated annotations for the same underlying entries. Use it when you specifically want RCSB's schema.
• AlphaFold Structures Scraper — predicted models (AlphaFold DB), not experimental depositions. Complementary, not overlapping: PDBe = experimentally determined; AlphaFold = computationally predicted.

If you need structure↔sequence↔function joins, pair this with the UniProt Scraper and InterPro Scraper.

Related actors

• RCSB PDB Scraper
• AlphaFold Structures Scraper
• UniProt Scraper
• InterPro Scraper

Notes

Data is sourced from the public PDBe REST API (EMBL-EBI). PDB data is released under CC0. Please cite PDBe/wwPDB in downstream work.