Drug-target interaction prediction method based on hypergraph neural network

Abstract

The invention provides a drug-target interaction prediction method based on a hypergraph neural network, and the method comprises the steps: firstly searching drug target interaction information froma public database as a data set, dividing the obtained data set into a training set and a test set, building a heterogeneous biological network for modeling through employing the data in the trainingset, modeling the heterogeneous biological network to obtain a heterogeneous biological hypergraph, generating a dual hypergraph of the heterogeneous biological hypergraph according to the obtained heterogeneous biological hypergraph, performing drug and target feature extraction by utilizing a hypergraph neural network, generating drug embedding and target embedding, and finally calculating the probability of interaction between a drug and a target. According to the technical scheme, the heterogeneous biological network is modeled into the hypergraph, the high-order complex relation between the medicine and the target can be fully learned, and a better prediction effect is brought.

Description

technical field [0001] The invention belongs to the field of computational biology, in particular to computational drug discovery methods, in particular to a drug-target interaction prediction method based on a hypergraph neural network. Background technique [0002] The identification of drug-target interactions (DTIs) is an important step in developing new drugs and understanding their side effects. Due to the increasing number of synthetic compounds developed to target a large number of proteins and disease processes, identifying drug-target interactions using biological assays is time-consuming and expensive, a major dilemma facing traditional drug discovery approaches. In recent years, to alleviate these shortcomings, researchers have attempted to use computational methods to identify drug-target interactions. [0003] The deep learning method in the computing method has been well applied in different fields, such as the application of convolutional neural network in t...

AI Technical Summary

Problems solved by technology

Therefore, methods based on network or ordinary graph models are difficult to capture the high-order complex relationship between drugs and targets.

Images