Simulation And Modeling Of The Relation Between SNP Multiple Pathogenic Factors And Diseases

With the development of high-throughput genotyping technology,the research focus on complex diseases has gradually shifted to genome-wide association analysis(GWAS),and the main target are single nucleotide polymorphism(SNP).SNP refers to a DNA sequence polymorphism caused by a single nucleotide variation in the genome,and is one of the most common heritable variations in human inheritance.Studies have shown that most of the diseases that humans suffer from are closely related to SNP,which can assist doctors in diagnosing and accurately treating patients in advance by modeling the causal relationship between multiple pathogenic factors and diseases.However,the SNP data set usually has a small sample size,low signal-to-noise ratio,and does not contain real information on the pathogenic model,which is not conducive to subsequent SNP interaction analysis,and the current research on the impact of different pathogenic sites on the disease is not enough Indepth,how to model the causal relationship between the pathogenic model and the disease is the key.In response to the above problems,the work and innovative results obtained in this research are as follows:1.Propose a multi-cause pathogenic data simulation method,which mainly combines the biological characteristics of the SNP data,the relationship between the SPN pathogenic model and the multi-pathogenic model to design,and according to the single-site pathogenic model and the multi-site model Two processes of pathogenicity model are implemented.The multi-cause pathogenic data simulation method designed in this study mainly addresses the causal relationship between multiple pathogenic causes and the disease.Compared with other simulation methods,this method allows users to set their own pathogenic models according to research needs,and can combine multiple pathogenic factors.It has the advantages of embedding a different pathogenic model into the data,setting a 2-way INME pathogenic model and not relying on random seeds,and being able to efficiently and quickly generate biologically characteristic multi-cause pathogenic SNP data.2.In the stage of modeling the causal relationship between multi-causal pathogenic factors and diseases,according to modern causality,a concept of sample membership is innovatively proposed,and a mathematical model is given to analyze diseased samples and pathogenicity Causality between models.3.Through sample membership,a membership-based pathogenic factor and disease association modeling algorithm MMA is proposed.This algorithm uses sample stacking for multi-cause disease problems,and each sample is based on the sample membership.Soft division is performed on each pathogenic model,and the causal relationship between the pathogenic model and the sample is updated through the results of the division,and stable results is obtained through continuous iteration.Experimental results show that the algorithm can accurately reflect the causal relationship between multiple pathogenic models and diseases.Compared with other algorithms,the accuracy of the algorithm is improved by more than 40%,and the algorithm does not need to set hyper parameters.do not rely on the results of the initial division,and is good at determining the advantages of multi-cause disease and disease causality.In summary,in view of the problem of multiple causes and disease causality,this paper designs a multiple cause disease data simulation system,defines sample membership,and proposes a membership based modeling algorithm for causality between pathogenic factors and disease MMA,which applies simulation data and real data to MMA,analyzes the MMA algorithm through experiments,and compares it with other algorithms,contributing to the analysis of complex diseases and the development of precision medicine.