|
Activity Number:
|
392
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 15, 2002 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #300788 |
|
Title:
|
Non-Traditional Statistical Approaches for the Analysis of High-Dimensional Genetic Data
|
|
Author(s):
|
Jason Moore*+
|
|
Affiliation(s):
|
Vanderbilt Program In Human Genetics
|
|
Address:
|
519 Light Hall, Nashville, Tennessee, 37232-0700, U.S.A.
|
|
Keywords:
|
DNA microarrays ; machine learning ; gene-gene interactions ; data reduction ; new methods ; pattern recognition
|
|
Abstract:
|
We are facing an information explosion and a comprehension implosion in the field of human genetics. New technologies, such as DNA microarrays, have made it feasible to simultaneously measure expression levels and DNA sequence variations for thousands of genes. Despite these technological advances, the development of statistical methods for evaluating the relationship between high-dimensional genetic data and biological and clinical endpoints has not kept pace. This is partly due to the complexity of the genetic effects that are to be modeled. There is an increasing awarness that high-order interactions among genetic factors will be more important than the independent main effects of each factor for many common clincial endpoints. Traditional methods such as logistic regression are not suited for modeling high-order genetic interactions because of limitations of sample size. We review several alternative approaches to modeling high-dimensional genetic data for common complex human diseases.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2002 program |
