RBP400 and RBP400-Expanded: Disease-aware protein candidate resources for ARISE-PPI

This record contains the RBP400 disease-aware protein candidate resource andthe RBP400-Expanded curated pair-level dataset used in the ARISE-PPI study. RBP400 contains standardized protein identifiers, disease-context candidateannotations and reference information for downstream protein-interactionprioritization and exploratory analysis.RBP400-Expanded contains curated positive and negative protein pairs,protein identifier mappings, sequence-cluster assignments, train-validation-testsplit manifests, negative-pair provenance records and excluded conflict records.The dataset was designed for controlled pair-level evaluation undercluster-disjoint partitions.Raw third-party databases are not redistributed in this record and remainsubject to their original licenses and access conditions.