Download Data

Export the PlasticDB dataset for offline analysis. Every file below is generated on demand from the live database: it holds every record that is public at the moment you download it.

Live, stamped with the current release. New organisms and proteins wait (staged) for the weekly release; new evidence on records that are already public, corrections, papers and vocabulary appear as soon as they are curated. Each file is named after the release it was built on (currently, e.g. plasticdb_proteins_2026.39.12.csv) and the response carries an X-PlasticDB-Data-Version header; dataset.json also states it in its metadata.release field and dataset.zip in its README.md. A file downloaded mid-week can therefore contain a little more than the release it names.

Custom filtered export

Export only the records you need — e.g. "PET degraders since 2020" — instead of the full dataset. Filters compose across all output formats.

Entity

Plastic

Include sub-types

Exact abbreviation or full name — PE does not match PET.

Year ≥

Substrate form

Evidence method The assay behind the evidence — not a quality ranking.

Include negative results

Min kcat (s⁻¹)

Max Km (mM)

Min kcat/Km (M⁻¹·s⁻¹)

CSV JSON FASTA

Per-entity exports

Microorganisms

CSV JSON

Proteins

Wild-type and engineered, tagged by a kind column. Add ?kind=wild_type for wild-type only.

CSV JSON FASTA

Plastics

CSV JSON

Papers

CSV JSON

Full dataset

Every entity and association in one bundle, engineered variants included — with the evidence spine (interactions, experiments, observations) as tables that join to the organisms, proteins, plastics and papers. The zip carries a README.md describing every column and a LICENSE.txt; dataset.json opens with the same information under metadata.

dataset.zip dataset.json

Engineered variants

Engineered enzyme variants with mutations, kinetics, and stability data. Join to the protein exports and the FASTA on seq_id.

CSV JSON

Dataset releases

PlasticDB releases its data weekly. Curators work continuously; new organisms and proteins they add become public only at the weekly release, which gives the data a citable number. Edits to records that are already public go live when they are saved, so the downloads above are the live data, named after the latest release.

The data on this site is release 2026.39.12. Release numbers are ISO year and week (2026.35 is the 35th week of 2026) and are separate from the application version in the footer: v1.32.0 says which build of the site you are using, the release says which data you searched.

When you report what you searched, cite the release — not the date you downloaded:

PlasticDB dataset release 2026.39.12. University of Auckland.

/api/v1/releases/current returns the current version, its Figshare DOI and a ready-made citation; /api/v1/releases is the full version history. Each release cuts an archived snapshot and mints a DOI only when the published data has actually changed, so a quiet week does not produce a second DOI for an identical dataset. Every archive names its own release inside README.md and snapshot.json; the live downloads above name theirs in the file name and the X-PlasticDB-Data-Version header.

Legacy static snapshots (PlasticDB v1 / pre-built)

degraders_list.tsv · PlasticDB.fasta · structures.tar.gz (predicted structures; the live database predicts with ESMFold)

Please Cite

If you use PlasticDB data in your research, please cite: Gambarini V, Pantos O, Kingsbury JM, Weaver L, Handley KM, Lear G (2022). "PlasticDB: a database of microorganisms and proteins linked to plastic biodegradation." Database, 2022:baac008. doi.org/10.1093/database/baac008

Every download is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0): reuse it freely, with credit. Licence details and how to cite a release.