Drug-Target Interaction Prediction Based on Attentive FP and Word2vec

The study of drug-target interactions (DTIs) plays a crucial role in localizing new drug targets, drug discovery, and drug improvement. Nevertheless, traditional DTI experimental determination is time-consuming and labor-intensive work. Therefore, more and more researchers are investing in using computational methods to predict drug-target interactions. This paper uses the neural network model to conduct further in-depth research on DTI. We use the Word2vec model to extract protein features, use the Attentive FP model to process drugs, obtain their features, and evaluate the results through several widely used model evaluation methods. The experiment result shows that our method works well on human and C.elegans datasets.