Generalized Disequilibrium Test For Association In Qualitative Traits Incorporating Imprinting Effects And Maternal Effect Based On Extended Pedigrees

BackgroundAssociation analysis is a useful tool to locate disease loci by using markers,which is widely used in case-control data and family data.If a marker locus is very close to a disease locus,linkage disequilibrium may exist between them.Association analysis utilizes the linkage disequilibrium to map the genetic variants by statistical methods based on the traits and genotypes.For a diallelic marker locus,there have been many family-based methods to test for the association between genotype scores and dichotomous traits.Among them,the generalized disequilibrium test(GDT)is a powerful method,which is the generalization of the traditional transmission disequilibrium test by using the genotype differences between all discordant relative pairs(including those beyond first-degree relatives)within a family.However,GDT cannot be used to incomplete family data directly.Genomic imprinting is an important epigenetic phenomenon in studying complex traits,where the expression levels of certain genes rely on their parental origin.If the disease allele transmitted by the father is imprinted and the one transmitted by the mother is expressed,it is paternal imprinting.On the contrary,if the disease allele transmitted by the mother is imprinted and the one transmitted by the father is expressed,it is maternal imprinting.i Furthermore,some researches have demonstrated that genomic imprinting plays an important role in several human genetic diseases such as Beckwith-Wiedemann syndrome,pseudohypoparathyroidism and transient neonatal diabetes mellitus.Currently,there has been increasing interest of incorporating imprinting to improve the test power of association analysis.As such,the pedigree disequilibrium test with imprinting(PDTI)and its extension Monte Carlo(MC)PDTI(MCPDTI)to accommodate pedigrees with missing genotypes were proposed to test for association,which consider the influence of imprinting on association study.However,they only utilize the genotype differences between all first-degree relative pairs in a family,which may reduce their test powers if ignoring the information on the genotype differences between beyond first-degree relatives.On the other hand,GDT does not take imprinting effects into account,and it has not been investigated whether it can be used for association analysis when the effects indeed exist.Another common epigenetic phenomenon is maternal effect.Maternal effect refers to the situation where the phenotype of an organism is determined by the genotype of its mother without focusing on the effect of its own genotype.In genetics,maternal effect occurs when an organism shows the phenotype expected from the genotype of the mother,irrespective of its own genotype,often due to the mother supplying mRNA or proteins to the egg.Based on the mechanism of action of maternal effect,a variety of diseases,especially those being related to pregnancy outcomes,have been found to be affected by maternal effect,such as childhood cancer spina bifida,schizophrenia and high blood pressure.Both genomic imprinting and maternal effect can cause the parent-of-origin patterns of phenotypic variation.Although genomic imprinting and maternal effect are distinct biological processes,they may lead to the same parent-of-origin patterns of phenotypic variation.Thus,incorporating maternal effect to improve the test power of the association analysis is increasingly worthy of being done.Here,at the loci on autosomes,we consider:(1)developing MCGDT to deal with incomplete pedigrees by using a MC sampling and estimation scheme;(2)proposing GDT incorporating imprinting effects(GDTI)and Monte Carlo GDTI for complete and incomplete family data,respectively;(3)developing GDT incorporating imprinting effects and maternal effect(GDTIM)and Monte Carlo GDTIM(MCGDTIM)for complete and incomplete family data,respectively.Methods(1)For incomplete pedigrees,using a MC sampling and estimation scheme to infer the missing genotypes given the observed genotypes in each pedigree,we extend GDT,and develop MCGDT.The simulated size can be used to verify the validity of the proposed method and the simulated power can show its advantage over other existing methods.(2)For complete pedigrees,to incorporate genomic imprinting into association,based on a novel decomposition of the genotype score of an individual according to the paternal or maternal source of an allele,we develop the GDTI test to test for association incorporating imprinting for complete pedigrees without missing genotypes.Then,using a MC sampling and estimation scheme to infer the missing genotypes given the observed genotypes in each pedigree,we extend GDTI and develop MCGDTI to deal with incomplete pedigrees,in which some individuals'genotypes are unavailable.The simulated size can be used to verify the validity of the proposed methods and the simulated power can show advantage of GDTI and MCGDTI methods over GDT method considering the genotype differences between all discordant relative pairs(including those beyond first-degree relatives)within a family and MCPDTI method incorporating imprinting effects and the genotype differences between all first-degree relative pairs in a family.The application to the rheumatoid arthritis(RA)dataset is also used to demonstrate the advantage of MCGDTI over other methods.(3)For complete pedigrees,by modeling the genotype of the mothers of individuals,we propose the generalized disequilibrium test for association incorporating imprinting and maternal effect(GDTIM).Then,we use a MC sampling and estimation scheme to infer the missing genotypes given the observed genotypes in each pedigree and develop the MCGDTIM method.The simulated size can be used to verify the validity of the proposed methods and the simulated power can show their advantage over other existing methods.Results(1)Under simulation settings with different sample sizes,genotype missing rates and pedigree structures,MCGDT controls the size well under the null hypothesis of no association.MCGDT can recapture much information of missing genotypes.(2)Under simulation settings with different sample sizes,genotype missing rates,pedigree structures and degrees of imprinting effects models,GDTI,MCGDTI and MCGDT control the size well under the null hypothesis of no association and no imprinting.As for the simulated powers,under complete and incomplete imprinting effect models,our proposed GDTI and MCGDTI methods by considering the information on imprinting effects and all discordant relative pairs outperform all the existing test statistics.Further,the existing GDT and the proposed MCGDT,although not constructed under imprinting,can be used for testing association even when the effects exist.(3)Various simulation studies are conducted to access the performance of the proposed methods and the existing test statistics with settings of different combinations of genomic imprinting effects and maternal effect,different sample sizes,pedigree structures and disease types(common disease and rare disease).Under the null hypothesis of no association,no imprinting and no maternal effect,the proposed GDTIM and MCGDTIM methods control the size well.In the presence of genomic imprinting and maternal effect or only the maternal effect being present,GDTIM and MCGDTIM have the obvious advantage over other methods.When the genomic imprinting effects exist but the maternal effect does not exist,MCGDTI is more powerful.If there is no epigenetic phenomena,compared to GDT,GDTIM,MCGDTIM and MCGDT have the similar power.Conclusion(1)To deal with incomplete pedigrees by GDT,we utilize a MC sampling and estimation scheme to extend GDT and propose MCGDT.According to the simulation results,MCGDT can recapture much of the missing information.Thus,we recommend MCGDT in practice.(2)Compared to GDT,based on a novel decomposition of the genotype score of an individual according to the paternal or maternal source of an allele,we develop the GDTI test to test for association incorporating imprinting for complete pedigrees without missing genotypes.In contrast to MCPDTI,GDTI makes use of the genotype differences between all discordant relative pairs,including beyond first-degree relatives.To deal with incomplete pedigrees,we extend GDTI and develop MCGDTI and MCGDT using a MC sampling and estimation scheme,in which some individuals' genotypes are unavailable.Simulation results indicate that GDTI,MCGDTI and MCGDT control the size well under the null hypothesis of no association and no imprinting.Because of considering the information on imprinting effects and all discordant relative pairs,our proposed GDTI and MCGDTI methods outperform all the existing test statistics and MCGDTI can recapture much of the missing information.The application to the RA dataset also demonstrates the advantage of MCGDTI over other methods.Further,in this thesis,we demonstrate that,the existing GDT and the proposed MCGDT,although not constructed under imprinting,can be used for testing association even when the effects exist.Thus,GDTI and MCGDTI are more practical.(3)In this study,we extend GDT and develop GDTIM by incorporating both genomic imprinting and maternal effect.Being similar to GDTI,we develop a novel decomposition of the genotype score of each individual according to the paternal or maternal source of the allele.For maternal effect,we additionally model the association between the disease status of individuals and his(her)maternal genotype scores.Furthermore,we further develop Monte Carlo GDTIM(MCGDTIM)for incomplete pedigree data where the genotypes of some members in pedigrees are missing by using MC sampling and estimation scheme.Simulation results indicate that GDTIM and MCGDTIM control the size well under the null hypothesis.In the presence of genomic imprinting and maternal effect or only the maternal effect being present,GDTIM and MCGDTIM have the obvious advantage.When the genomic imprinting effects exist but the maternal effect does not exist,compared to MCGDTI,MCGDTIM considers the wrong information on maternal effect,thus,MCGDTI is more powerful.However,because of considering the correct information on imprinting effects,MCGDTIM outperforms MCGDT.