CATH v4.4.0 protein domain classification with an S35-aware train/test split

Provider
CATH (Orengo group, UCL); packaged by LiteFold
License
CC-BY-4.0 (source)
Size
16 files
Format
Parquet (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/LiteFold/CATH)
Access
Free and public. No account with the data provider is needed.
Category
Structural Biology
Source
https://www.cathdb.info/

About CATH v4.4.0 protein domain classification with an S35-aware train/test split

The CATH protein domain classification (class, architecture, topology, homologous superfamily) for 601,328 domains, packaged by LiteFold with the original CATH v4.4.0 release files and a machine-learning split. The original files include cath-domain-boundaries.txt, the domain lists at S35, S60, S95 and S100 sequence-identity levels, and the non-redundant S40 domain structures as a PDB tarball. The Parquet split puts 541,123 domains in train and 60,205 in test, keeping every domain of the same homologous superfamily and S35 cluster on one side, so close relatives never cross the split. It covers 6,630 superfamilies, 1,472 topologies and 43 architectures, with sequence lengths from 9 to 1,275 residues. Use it to train or benchmark fold and function classifiers. CATH is CC BY 4.0.

How to load this dataset in Python

from datasets import load_dataset
ds = load_dataset("LiteFold/CATH")
print(ds)

Data structure

File types in sampled files: Parquet

Files

Example file: data/test-00000-of-00001.parquet

Analyze this dataset without downloading it

On Scigantic, CATH v4.4.0 protein domain classification with an S35-aware train/test split opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Structural Biology datasets