DTI

The dataset is divided into two parts. Each dataset contains the following components:mat\/ folderThis folder stores the association matrices between different types of entities. Each matrix represents the known relationships (e.g., drug\u2013target, disease\u2013microbe) for the corresponding dataset.similarity\/ folderThis folder includes the similarity matrices between entities of the same type (e.g., drug\u2013drug similarity, target\u2013target similarity). These matrices are typically computed based on features such as chemical structure, genomic sequence, or functional annotation.processed\/ folderThis folder contains the preprocessing scripts used to convert the raw data into a format suitable for model training and evaluation. These scripts handle data normalization, graph construction, and train-test splitting.5_folds\/ folderThis folder contains the five-fold cross-validation datasets. Each fold consists of a distinct training and testing split to ensure robust evaluation of model performance.