Graph Convolutional Network for Drug Response Prediction Using Gene Expression Data

Genomic profiles of cancer patients such as gene expression have become a major source to predict responses drugs in the era personalized medicine. As large-scale drug screening data with cell lines are available, number computational methods been developed for response prediction. However, few incorporate both and biological network, which can harbor essential information about underlying process response. We proposed an analysis framework called DrugGCN prediction Drug using Graph Convolutional Network (GCN). first generates graph by combining Protein-Protein Interaction (PPI) network feature selection drug-related genes, GCN model detects local features subnetworks genes that contribute localized filtering. demonstrated effectiveness showing its high accuracy among competing methods.