PRING Dataset
收藏DataCite Commons2025-05-16 更新2025-05-17 收录
下载链接：
https://dataverse.harvard.edu/citation?persistentId=doi:10.7910/DVN/22AUPR
下载链接
链接失效反馈官方服务：
资源简介：
The PRING dataset provides a comprehensive benchmark for protein-protein interaction (PPI) network prediction and reconstruction. It includes curated PPI graphs from four model organisms, Human, Yeast, Ecoli, and Arath, collected from STRING, UniProt, IntAct, and Reactome databases. The dataset is split into training and test sets under a leakage-free protocol to ensure fair evaluation.
PRING goes beyond binary classification by supporting a variety of topology- and function-level evaluation tasks, including:
Graph topology reconstruction (e.g., graph similarity, clustering coefficient)
Functional module detection and GO enrichment analysis
Cross-species generalization assessment
Essential protein identification
This dataset enables rigorous and biologically meaningful evaluation of machine learning models for PPI prediction and network biology.
提供机构：
Harvard Dataverse
创建时间：
2025-05-16
