Supporting datasets for DGMAP disease–gene association prediction

This dataset contains the data used for the experiments and case study in the DGMAP disease–gene association prediction study. It includes the processed DisGeNET-v1 and DisGeNET-v2 benchmark datasets used for model training and evaluation, as well as five Gene Ontology annotation files manually retrieved from QuickGO for the Alzheimer’s disease gene-network analysis. The benchmark datasets were obtained from the publicly available FactorHNE and HOGCN-LP repositories, respectively. The original benchmark files were converted into NumPy format for computational convenience without modifying their biological associations or content. Detailed provenance and preprocessing information are provided in the accompanying README file.