Method for constructing gene regulatory network based on structural prediction

Abstract

The invention provides a method for constructing a gene regulatory network based on structural prediction. The method includes the following steps: first calculating a coefficient matrix, determiningthe correlation between genes, by calculating the Peason coefficient, the mutual information, and the maximum mutual information between the genes, as a basis for screening potential parent node sets;then, performing structural prediction, using an obtained coefficient matrix between genes as the basis for determining the potential parent node sets of the genes, and selecting a potential parent node set for each gene; finally, performing the structural learning and parameter learning of the gene regulatory network. The method predicts the potential parent node sets of the genes by a method based on of the combination of the Person coefficient, the mutual information, and the maximum mutual information, narrows the search range of structure learning, shortens the construction time of the gene regulatory network to a certain extent, improves computing performance, can build a large-scale gene regulatory network quickly and accurately, and further understands the gene regulatory mechanism of organisms.

Description

technical field [0001] The invention relates to the field of medical informatics, in particular to a method for constructing a gene regulatory network based on structure prediction. Background technique [0002] Gene expression regulation is the process in which the expression of a gene is affected by other genes, mainly including regulation at the level of transcription and translation. Since the final form of gene expression is protein, it needs to be transcribed to form mRNA as a template to finally produce protein, so gene regulation at the transcriptional level is the key. The transcription of a gene is affected by the expression products of other genes, which play a stimulating or inhibiting role, and the proteins produced by itself may also affect other genes. This complex regulatory relationship ultimately constitutes a gene regulatory network. Understanding the gene regulation mechanism of organisms can understand the occurrence of various biological processes fro...

AI Technical Summary

Problems solved by technology

For example, the Boolean model is too rough and simplified to study the gene regulatory network from a qualitative point of view; the differential equation model quantitatively and accurately describes the gene regulatory network through differential equations, but at the same time, it will cause problems that are difficult to optimize due to too many parameters, and the amount of calculation huge

The Bayesian network model describes the gene regulatory network through a probability model, and uses probability to represent the regulatory relationship. However, as the complexity of the network increases, the amount of calculation of the model will increase significantly.

Images