tcbb2014

tcbb2014数据集收集自DrugBank，KEGG Drug，DCDB和Matador，包括315种药物，250种靶标，5种药物之间的相似性类别和3种靶标之间的相似性类别。药物之间的相似性包括在化学，配体，性能，副作用和注释方面的相似性。靶标之间的相似性包括序列，蛋白质-蛋白质相互作用网络和基因本体相似性。该数据集的原始任务是根据网络中不同类型的相似性来预测药物与靶标之间的新型相互作用

The tcbb2014 dataset is curated from four public databases: DrugBank, KEGG Drug, DCDB and Matador. It contains 315 drugs, 250 drug targets, 5 drug similarity categories and 3 target similarity categories. Drug similarity encompasses five categories: chemical structure similarity, ligand similarity, pharmacological property similarity, side effect similarity and annotation similarity. Target similarity includes three categories: sequence similarity, protein-protein interaction (PPI) network similarity and gene ontology (GO) similarity. The original task of this dataset is to predict novel drug-target interactions based on various types of similarity information within the network.