MEASURING AND USING INDIVIDUAL GENOMIC ANCESTRY TO STUDY COMPLEX PHENOTYPES
Open Access
- Author:
- Halder, Indrani
- Graduate Program:
- Genetics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- August 08, 2005
- Committee Members:
- Mark Shriver, Committee Chair/Co-Chair
Kenneth Monrad Weiss, Committee Chair/Co-Chair
Hiroshi Akashi, Committee Member
Kateryna Dmytrivna Makova, Committee Member
Esteban J Parra, Committee Member - Keywords:
- Genomic ancestry
individual ancestry estimation
population stratification
admixture mapping
complex diseases
racial/ethnic differences in disease risk - Abstract:
- Several complex diseases show population specific differences the causes for which are as yet unknown. Examples of such diseases include obesity and non insulin dependent diabetes which are more prevalent in African American, Indigenous American and Latino populations compared to European Americans. Dementia and Osteoporosis are examples of diseases which are more prevalent in European Americans. It is obvious from these examples that the differences follow broad racial or ethnic categories and because of this correlation, medical research has focused extensively on investigating racial or ethnic differences in disease risk. The concept of ‘Race’ is multifaceted and any simplistic annotation is insufficient, may be even wrong, unless we make attempts to decompose the different aspects of race and use them appropriately in the context of biomedical research. Broadly speaking, race has two distinct components. The biological component referred to as Biogeographical ancestry; and the sociocultural component which reflects the ethnic heritage of an individual. Anthropological and genetic research in the past couple of decades has shown that the concept of typological races is scientifically obsolete yet the idea lives on and is used in biomedical research. Race/Ethnicity is still used as a classifier to establish and highlight population specific differences, while often inter-individual variation within a specific racial/ethnic category is ignored. Anthropological research has shown that anatomically modern humans originated in Africa, and migrated to different parts of the world. Over the course of time, continental populations were established and for a period, large scale interactions between continental populations were limited. Given the recent and common origin, most genomic regions in all continental populations are very similar. At some loci, however, there has been some change in allele frequencies in the time since the separation of populations. In the past few centuries, there have been large scale migrations, voluntary or forced, that gave rise to populations of “mixed” ancestry. And at those loci where allele frequencies differed between continental populations, this has resulted in long range non random association in the admixed population. Many US residents can trace their genetic ancestry to more than one continent. The European colonial period that started in the late 1400s brought together in the New World, populations that had been geographically isolated, namely, Europeans, West Africans and Indigenous Americans. The impact of this was two fold, biological, which resulted in genes from different ancestral populations coming together in varying proportions in the different admixed populations and also cultural, in that certain ways of life of one population were adopted by another. Thus, the differences in disease prevalence which appear to follow broad racial categories could have been influenced by both genetic and non genetic factors (including environmental, sociocultural and behavioral factors). Several resident US populations including European Americans, African Americans and Mexican Americans, which are classified as different races in the biomedical literature have been studied in this thesis. Populations with a recent history of admixture provide a unique opportunity for understanding the genetic bases of many common diseases. If ancestry of alleles at each locus can be established, then it will be possible to correlate the proportion of alleles inherited from an ancestral population to a phenotype and thus establish a genetic basis for the disease. This will help in decomposing the different sources (genetic and non-genetic) that contribute to the phenotype. Thus, estimating ancestry of alleles at a locus is important. Several factors influence the estimate, including nature and number of markers used, the method used for estimating ancestry and the model of admixture that is assumed. Mitochondrial and Y chromosome segments only provide information regarding paternal or maternal lineage. Autosomal loci on the other hand represent all ancestors of an individual and these are the loci that have been studied in this thesis. The distribution of individual genomic ancestry in different populations has been estimated using several different methods and it is shown that wide variations exist among individuals who self identify as being from the same racial or ethnic category. In addition, significant overlap is seen between two of the racial categories, European Americans and Latinos which are described in this thesis. Modest amounts of geographic variation are also observed within the same racial category. Additional simulation studies have been done to investigate the effects of markers, model specifications and method choice on the reliability of the estimates of individual genomic ancestry obtained. Admixture mapping is an approach that uses the ancestry of alleles at a locus in mapping genes for diseases that differ among populations. Several methods have been developed recently to use this approach. This method is suitable for complex diseases since it is unaffected by locus heterogeneity and allelic heterogeneity. In addition unknown gene-environmental interactions further complicate the study of such diseases. Preliminary admixture mapping of hypertension and obesity have been performed and are presented. One locus has been identified that shows strong linkage with obesity in an African American sample.