Gene selection method and system based on adaptive gene interaction regularization elastic network model

Abstract

The invention discloses a gene selection method and system based on a self-adaptive gene interaction regularization elastic network model. The method comprises the following steps: checking and evaluating the importance degree of each measured gene based on a Wilcoxon rank sum; quantifying the importance degree of each measured gene, adding an adaptive penalty weight, and then deleting noise genes to obtain feature genes; introducing the penalty weight into a least square loss function, and constructing a self-adaptive elastic network model; constructing an adjacent matrix of the gene interaction network; constructing a gene interaction network penalty based on the adjacent matrix; combining the self-adaptive elastic network model with gene interaction network penalty to construct a self-adaptive gene interaction regularization elastic network model; solving an optimal solution of the regularization elastic network model based on a gradient descent algorithm, and selecting a gene based on the optimal solution. According to the method, important genes highly related to tumor generation can be adaptively selected, and redundant and unrelated genes and noise genes are removed.

Description

technical field [0001] The invention belongs to the technical field of biological information, and in particular relates to a gene selection method and system based on an adaptive gene interaction regularized elastic network model. Background technique [0002] Tumor has now become one of the major diseases that threaten human life and health. According to the 2018 Global Cancer Statistics Report, the number of new cancer cases in the world will reach 18.1 million in 2018, and the number of deaths will reach 9.6 million. It is still rising rapidly, but the research on the treatment and prevention of cancer is not very comprehensive. With the application and development of a large number of gene chip technology, human beings have been able to obtain the normal gene expression information of various tissues one after another, and find out a small number of genes with differential expression in different disease categories from the large number of genes measured on the gene chi...

AI Technical Summary

Problems solved by technology

[0011] Aiming at the problems of low estimation efficiency, insufficient consideration of gene interaction and relatively large task load existing in existing gene selection methods, the present invention proposes a gene selection method and system based on an adaptive gene interaction regularized elastic network model, which overcomes the The above defects can adaptively select important genes that are highly related to tumor generation, and remove redundant, irrelevant genes and noise genes

Images