|
Activity Number:
|
64
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #306029 |
|
Title:
|
An Importance Sampling Procedure for Obtaining Confidence Intervals of Disease Loci with General Pedigree Data
|
|
Author(s):
|
Shuyan Wan*+ and Shili Lin
|
|
Companies:
|
The Ohio State University and The Ohio State University
|
|
Address:
|
Cockins Hall, Room 404, Columbus, OH, 43210,
|
|
Keywords:
|
importance sampling ; pedigree data ; CSI ; LOD
|
|
Abstract:
|
To localize a disease gene in a whole-genome study, we extended the confidence set approach idea of Lin et al.[2001] to general pedigree data assuming a known disease model. We test every position on the chromosomes based on the LOD score statistic. Those positions not rejected form a confidence interval for the disease locus. Three approaches were proposed to perform the tests, including a crude procedure for testing every fixed distance, a refined procedure based on inference for LOD score and an importance sampling (IS) procedure. Performance of the approaches was evaluated based on simulation studies. The IS procedure stands out in terms of the power and computing efficiency. Asymptotic properties of the null distribution was also examined. We will compare the performance of the asymptotic approximation with that of the simulation methods and apply our methods to a real data set.
|
