Method for decomposing m6A methylation local function spectrum fused with genome characteristics

Abstract

The invention relates to a method for decomposing m6A methylation local function spectrum fused with genome characteristics, and belongs to the field of epitranscriptomics and pattern recognition research. The invention aims at performing high-precision decomposition on the RNA methylation local function spectrum by fusing genome characteristics and revealing a detailed regulation mechanism of m6A methylation. Based on an independent component analysis (ICA) method in matrix decomposition, genome characteristics corresponding to m6A methylation sites are fused so as to fully consider relevance between extremely strong correlation m6A sites. Finally, a plurality of regulatory pathways in which the m6A locus participates are recovered by using a negentropy-based estimation method, and a sensitive locus corresponding to each regulatory pathway is further mined to realize high-precision spectral decomposition of the m6A methylation local function. The algorithm provided by the invention can provide powerful reference for revealing a regulation mechanism of m6A methylation for wetland experiments.

Description

technical field [0001] The invention relates to the field of epitranscriptomics and pattern recognition research, in particular to an m 6 A Methylation Local Functional Spectrum Decomposition Method. Background technique [0002] In epitranscriptomics, N6-methyladenine (m 6 A) As the most common and abundant post-transcriptional RNA modification in eukaryotic mRNA, it refers to the methylation that occurs on the sixth N atom of base A. Although many recent studies have shown that m 6 A modification is related to key biological functions and pathological phenomena, but m 6 The detailed regulatory mechanism of A methylation is still unclear. Multiple studies have shown that each m 6 A methylation regulatory factors all regulate the methylation level of many sites at the same time. The methylation levels of loci that share the same regulatory factor are generally consistent with changes in the regulatory factor, and are significantly correlated functionally, showing a coo...

AI Technical Summary

Problems solved by technology

However, on the one hand, the existing methods based on the idea of ​​matrix decomposition have only been tested and analyzed on the gene expression profile, and have not actually optimized or improved the RNA methylation profile; on the other hand, the RNA methylation profile The implicit information included (such as the genomic characteristics of the site, etc.) has not been fully utilized, which leads to a certain improvement in the spectral decomposition accuracy of many current methods.

Images