SUMMARY To investigate the genetic basis of age-related hearing impairment (ARHI) and tinnitus we will analyze genotype array, whole genome imputed and exome sequence data from 500,000 participants in the UK Biobank. We will conduct single and rare variant aggregate association tests: testing for main effects, sex and age specific associations and interactions (gene x gene; gene x environment; gene x age; gene x age; and gene x sex) controlling for important confounders, e.g. noise exposure. We perform fine mapping to tease apart functional causal variants from those which are in linkage disequilibrium. We will also test for pleiotropy and perform mediation analysis to determine if biological or mediated pleiotropy has been detected. Additionally, we plan to develop novel approaches to analyze imputed genetic data that explicitly account for the uncertainty in genotype calls during association analysis. By ignoring or improperly modeling the uncertainty in imputed genotypes, current methods suffer from a decreased ability to detect associations as well as an increased false positive findings rate. Therefore, we will develop methods to analyze imputed data, which properly models imputed genotype data uncertainty to allow for the detection of associations, interactions, pleiotropy and fine mapping. The novel methods will be thoroughly evaluated and implemented in our SEQSpark software to perform data quality control, annotation, and association analysis for hundreds of thousands of samples with imputed genotype data. This study has the potential for significant public health impact by providing a useful analytic tool to the research community and by conducting a well-powered, comprehensive investigation of the genetic etiology of ARHI and tinnitus which in turn will aid in risk prediction, prevention, and improved and new treatment modalities.