Home > Events > 2016 Seminars & Colloquia > Xiang Zhan, Fred Hutchinson Cancer Research Center

Xiang Zhan, Fred Hutchinson Cancer Research Center

Main Content

Measuring and testing independence by KRV coefficient with application to genome-wide association studies
When
11 February 2016 from 4:00 PM to 5:00 PM
Where
201 Thomas Building
Add event to calendar
vCal
iCal

Xiang ZhanMeasuring and testing independence by KRV coefficient with application to genome-wide association studies

 

 

Genome-wide association studies (GWAS) have been established as an important tool to identify genetic variants associated with many complex traits and diseases. Recent advances in high-throughput biotechnology have revolutionized biological and biomedical research. An unprecedented wealth of data generated from those studies has created great opportunities and challenges for researchers. One of such challenges considered in this paper is the association analysis between multiple traits and multiple genetic variants. To address this issue, we propose a kernelized RV (KRV) coefficient as a new measure of dependence between two random vectors. Under mild regulatory conditions, we show that the proposed KRV coefficient is zero if and only if two random vectors are independent. By studying the distribution of the KRV coefficient under the null hypothesis of independence, we build a KRV-based independence test, which is further tailored for testing association in GWAS. The proposed KRV test is shown to have corrected type I error rate and higher power than other existing methods in both simulation studies and a gene pathway expression data.

 

Filed under: ,

Navigation for this Section

Events