Font Size: a A A

Heritability Of Nicotine Dependence Based On Linear Mixed Model

Posted on:2020-06-19Degree:MasterType:Thesis
Country:ChinaCandidate:T T LanFull Text:PDF
GTID:2370330590955940Subject:Public health
Abstract/Summary:PDF Full Text Request
Objective: Tobacco use is the largest preventable cause of death in the world today.The prevalence of tobacco use not only causes harm to human health,but also causes serious economic burden to the society.Nicotine is the main harmful substance in tobacco.The greatest harm is its addiction.Smokers will continue to smoke,resulting in nicotine dependence.Nicotine dependence is a complex multi-dimensional phenotype,which is jointly affected by genetic and environmental factors.Heritability is used to measure the effects of genetic factors on phenotype.It is widely used by researchers.Numerous family and twin studies have proved a significant genetic influence on nicotine dependence,but there are often confounding environmental factors.The emergence of high-throughput sequencing technology enables us to directly estimate the heritability related to SNPs.Many SNPs related nicotine-dependent have been identified by GWAS,but the SNPs identified by GWAS account for only a small proportion of the heritability.In this study,genome-wide SNPs were used to more accurately estimate the effects of genetic factors on nicotine dependence to find more SNP sites causing nicotine dependenceMethods: The data were whole-genome sequencing data and phenotypic values from 1,405 individuals and from 681 families in the Minnesota twins and families study sequencing sample.The model used in this study was linear mixed model(LMM),which was simulated for 4 models.For the first model,genome-wide SNPs and kinships were used as random effects,and gender was a covariate to add as fixed effects.The number of variance components is 2.For the second time,genome-wide SNPs were divided according to chromosomes,SNPs in each pair of chromosomes and the kinships were used as the random effects,and gender as the fixed effect.In this case,the number of variance components is 23.For the third time,genome-wide SNPs were divided according to the MAF,and the SNPs in each MAF group and kinships were used as random effects and gender as fixed effect.In this case,the number of variance components is 8.For the fourth time,genome-wide SNPs were divided according to the functional categories of genes,and the SNPs in each category and were used as random effects and gender as fixed effect.In this case,the number of variance components is 7.The variance of each component of all models was estimated by using Multi-component Restricted Maximum Likelihood(REML-MC)estimation,and then computed the heritability of each component.Simulation and estimation for heritability were done in GCTA software.Results: After analysis,the heritability of whole genome SNPs accounted for 44 percent of variations related nicotine dependence.When the whole genome SNPs is divided according to chromosomes,the heritability of each chromosome is obtained.Chromosome 8 had the highest heritability,accounting for 12 percent of the variations.chromosome 10 had also high heritability that accounted for 10.6 percent variations.Whole genome SNPs were divided according to the size of MAF to obtain the heritability of each group of MAF.Among them,the heritability was the highest when the minimum allele frequency was between 0.001 and 0.01,which account for 24.8 percent of variations.The whole genome SNPs were divided according to the functional categories of genes,and the heritability of SNPs of each functional category was obtained.Among them,the promoters of SNPs account for 38.1 percent of variations associated nicotine dependence.Conclusion: The estimation of heritability for genome-wide SNPs can more accurately represent the association between genetic factors and nicotine dependence.The genetic information related to nicotine dependence focuses relatively on chromosome 8,chromosome 10 and promoter positions,and maybe there are more significant effects in rare variations.Therefore,the further study can be focused on chromosomes8,chromosome10 and promoter to find more sites associated with nicotine dependence.Moreover,the role of common variants cannot be studied independently,because rare variants also have influence on the nicotine dependence.
Keywords/Search Tags:Whole Genome, Linear mixed model, Heritability, GCTA, Nicotine dependence
PDF Full Text Request
Related items