High-dimensional data feature selection method based on graph neural network and spectral clustering

Abstract

The invention provides a high-dimensional data feature selection method based on a graph neural network and spectral clustering. The method comprises the steps: taking each gene as a node to establisha gene relation graph structure model, taking gene correlation data as side information to be added into a gene relation graph, taking a graph neural network model for obtaining feature vector representation of the nodes, and after the feature vector representation of each node is obtained, starting a link prediction stage, generating a new edge based on the genetic relationship graph, and obtaining a new genetic relationship graph; finally, selecting a node with the highest weight from the new genetic relationship graph based on spectral clustering to serve as a feature node. According to the invention, the finally selected gene has small redundancy, a good model effect is achieved, and the interpretability of the biological angle is supported.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a high-dimensional data feature selection method based on graph neural network and spectral clustering. Background technique [0002] In the field of bioinformatics, most of the processed objects are multi-feature, high-noise, and non-linear data sets. For example, researchers can use gene chips to simultaneously detect the expression values ​​of thousands of genes in one experiment, thereby obtaining a large amount of gene expression data; they can also use protein mass spectrometry to produce a large amount of protein expression profile data at one time. However, due to the high dimensionality and small number of samples of these data, conventional pattern recognition methods are no longer applicable. For this kind of data, how to eliminate redundant features and mine hidden useful biological information from massive data has become the key to research on identificati...

AI Technical Summary

Problems solved by technology

However, these algorithms generally assume that the samples are independent of each other, or only consider the characteristic relationship between the data, and are limited to finding the best explanation at the statistical level in the limited sample space, focusing on the model results and ignoring the biological factors. or protein-protein interactions

Images