Gene regulation and control network inference method combining sparse regression and elimination rule

Abstract

The invention provides a gene regulation and control network inference method combining sparse regression and elimination rule. The gene regulation and control network inference method comprises the following steps: reading gene expression data; respectively establishing sparse regression models of the gene expression data; obtaining the weight and external noise of each regulation gene accordingto the sparse regression models, and establishing a weight matrix of all regulation genes; and implementing a weight elimination rule on the weight matrix, removing the regulation genes with relatively small weight absolute values according to the elimination rule, and constructing an inter-gene gene regulation and control network for the remaining genes in the gene expression data. According to the method, the gene regulation and control network can be quickly inferred, and the specific regulation and control relationship between genes can be determined. By combining an advanced machine learning algorithm and an optimization rule, an accurate and efficient mathematical model is constructed to deduce a gene regulation and control network from gene expression data.

Description

technical field [0001] The invention relates to the field of gene regulation, in particular to a gene regulation network inference method combining sparse regression and elimination rules. Background technique [0002] With the development of high-throughput sequencing technology, a large amount of gene expression data provides a reliable basis for research. The purpose of inferring gene regulatory network is to obtain the network structure composed of gene and gene mutual regulation from gene expression data, and then through the analysis of expression level and gene regulation relationship, it can be used to identify disease-causing genes, so as to provide guidance for the treatment of diseases. refer to. Although many inference methods already exist, using gene expression data to infer gene regulatory networks remains a major challenge due to the complex regulatory relationships among genes. [0003] Researchers have conducted many studies on the study of gene regulator...

AI Technical Summary

Problems solved by technology

The main manifestation is that it can only infer whether there is a regulatory relationship between genes, and it is impossible to determine whether the relationship is activation or inhibition. Inferring from a large-scale gene regulatory network, the calculation complexity is high and the operation speed is slow, and it cannot provide accurate information for identifying disease-causing genes. gene regulatory network

Images