Statistical methods for analyzing high-throughput genotype and sequencing data. Recent advances in sequencing technologies make it possible to sequence a large number of subjects and test many low frequency or rare genetic variants. Many new statistical methods have been developed to analyze sequencing data and the majority of these methods focus on population based samples. It is well- known that rare variants segregate in families and population based methods will suffer lower power when studying rare variants. In addition, how to control population stratification has not been well studied for rare variant analysis. As such, this renewal application requests support to continue developing statistical methods for analyzing large scale genetic data sets. In particular, we propose to 1) Develop unified statistical methods to detect rare genetic variants from sequencing data using both families and unrelated subjects;2) Develop statistical methods of joint modeling of local ancestry and association in rare variant analysis, at the same time controlling for population stratification;3) Develop statistical association methods that can be applied to multiple traits using summary statistics from GWAS or whole sequencing data;4) Develop corresponding software that will be made publicly available. We will collaborate with laboratory-based investigators to obtain appropriate data sets and apply our new analytic methods to this crucial practical problem in genetic epidemiology. In particular, we will be analyzing the GWAS and sequencing data on traits related to hypertension and sleep apnea.
The propose of this application is to develop new statistical methods for analyzing high- throughput genotyping and sequence data. We will develop statistical methods to detect rare genetic variants using both family and unrelated subjects, to jointly model admixture mapping and association in order to search for rare potential causal variants contributing to the admixture mapping signals and to analyze multiple traits using summary statistics from GWAS or whole sequencing data. Finally we will develop corresponding software to these new methods.
Showing the most recent 10 out of 64 publications
