Principal Investigator: Professor Jianzhi Zhang
University of Michigan, Ecology and Evolutionary Biology, 830 N. University, 1075 Natural Science Building, Ann Arbor, MI 48109, United StatesTags: 27800, effect, Epistasis, selection
1a: Diminishing returns from advantageous mutations refers to the phenomenon that the same advantageous mutation is less beneficial when occurring in fitter genotypes. This has never been studied on non-fitness phenotype. Based on our preliminary study in yeast, when a phenotype is under directional selection, a mutation that changes this phenotype toward the favored direction tends to have a smaller effect when occurring in genotypes with favored phenotypic values. The phenotype and genotype data in UK biobank data would allow us to test our prediction thoroughly, and the results will help us infer human phenotypes that are under directional selection.
1b: Understanding the effects of mutations is important for health-related studies. Our method will help reveal phenotypes under directional selection in humans, and how mutations interact with phenotypes, which is important for finding new methods to alleviate the effect of disease-causing mutations. The proposed research meets the purpose-“improve the prevention, diagnosis, and treatment of illness and promotion of health throughout society” of UK Biobank.
1c: We will first choose a range of quantitative disease/physiology phenotypes(such as weight, height, blood pressure, baldness, the number of biological/adopted siblings/kids, lifespans of parents, and etc), and either map the GWAS loci for each phenotype ourselves or search for the published GWAS loci. For each phenotype, we will divide the individuals into a high-value group and a low-value group. We will calculate the phenotypic effect of each GWAS locus from the two groups. We predict the direction of selection based on whether GWAS loci have biased larger/smaller effects in the high phenotype group.
1d: We need the full cohort to be included.
Identical by descent (IBD), shared chromosomal segments among individuals in a population, helps reveal relationships status among study samples. IBD information helps genetic association studies by enabling imputation of genetic variants, refining genetic association testing through explicit adjustment of relatedness. We develop new efficient algorithms for inferring IBD segments, that is feasible for biobank-scale samples. We will apply our methods and existing methods to the UK Biobank genotype data. We will conduct IBD-mapping to identify regions with high levels of IBD among carriers of a disease which may suggest potential causal genes. We will correlate IBD information with home location information to gain understanding of population dynamics. We will also correlate IBD information with a comprehensive list of phenotypes including baseline measurements and retina fundus and OCT images.