MiRNA-gene relationship prediction method and system based on deep learning heterogeneous information network

Abstract

The invention discloses a miRNA-gene relationship prediction method and system based on a deep learning heterogeneous information network. The method comprises the following steps: firstly, constructing a miRNA-gene heterogeneous information network; then collecting a path instance set between the miRNA-gene pair on the heterogeneous network by using a meta path, and capturing effective information of the path set by using a deep convolutional neural network; and finally, splicing miRNA embedding, gene embedding and path embedding, and predicting the interaction between miRNA and the gene through a multi-layer perceptron. According to the invention, the defect that the traditional machine learning needs to manually collect features is avoided, and the network node features are automatically learned by using a deep learning method in a form of network nodes. A contrast experiment result shows that the performance of the MDCNN is superior to that of other methods, and the potential miRNA-gene interaction can be accurately predicted.

Description

technical field [0001] The invention belongs to the technical field of computer biological information network embedding and machine learning, and in particular relates to a miRNA-gene relationship prediction method and system based on deep learning heterogeneous information network. Background technique [0002] MicroRNAs (miRNAs), as one of the most important components in cells, can cause gene degradation or inhibit gene translation by complementary pairing with 3'UTRs of mRNA. Biological experiments have confirmed that miRNAs are widely involved in a large number of cellular processes and are closely related to the occurrence and development of diseases. Studying the association of miRNA target genes is of great significance for understanding the function and regulation mechanism of miRNA and preventing and treating human diseases. Thanks to the continuous advancement of information technology, computer-aided miRNA-gene relationship prediction provides a powerful boost ...

AI Technical Summary

Problems solved by technology

Although in the past period of time, many predictive technologies based on matrix decomposition and traditional machine learning for miRNA target gene interaction have been proposed, but there is a common problem of relying on tedious manual feature data extraction, and the accuracy needs to be improved

Images