Fig. 1
From: A novel method for drug-target interaction prediction based on graph transformers model
The DTI-GTN method’s flow chart. DTI-GTN includes a feature extractor based on heterogeneous networks, a feature selector based on principal component analysis, a converter to transform drug-protein interaction maps into drug-target pairwise line maps, and a fully connected layer classifier based on the GTN model. First, using the Jaccard similarity coefficient and RWR algorithm, features were retrieved from a network of seven medicines and proteins. Following that, a PCA model is used to reduce the dimensionality of these features, as well as to turn the drug-protein interaction graph into a drug-protein pairwise line graph. Finally, a GTN model was developed to forecast the interactions of each drug-protein pair