MPLID: Membrane Protein-Lipid Interface Dataset v2.0.0
收藏Zenodo2026-02-06 更新2026-05-26 收录
下载链接:
https://zenodo.org/doi/10.5281/zenodo.18508696
下载链接
链接失效反馈官方服务:
资源简介:
A large-scale dataset of experimentally validated lipid contact residues derived from crystallographic structures in the Protein Data Bank.
100% CRYSTALLOGRAPHIC LABELS - NO COMPUTATIONAL DATABASE DEPENDENCIES
Dataset Statistics (v2.0.0)
Proteins: 7,541
Total residues: 10,486,965
Contact residues: 98,101
Contact rate: 0.94%
Sequence clusters: 1,848 (30% identity)
Lipid codes recognized: 121
Train/Validation/Test Splits
Train: 4,644 proteins, 7,521,766 residues
Val: 1,436 proteins, 1,507,429 residues
Test: 1,413 proteins, 1,391,166 residues
Key Features
Labels derived 100% from PDB crystallographic coordinates
4.0 Å Cα-to-lipid distance cutoff
140× more proteins than DREAMM benchmark
Cluster-aware splits prevent data leakage
Fully reproducible from public PDB data
GitHub: https://github.com/omagebright/MPLID
提供机构:
Zenodo创建时间:
2026-02-06



