The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Abstract Details
Activity Number:
|
190
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 1, 2011 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract - #302407 |
Title:
|
A Resampling-Based Approach for Controlling Type I Error in a Genomic Microarray Experiment in Which Case/Control Status and Chip Effect Are Confounded
|
Author(s):
|
Stephen Erickson*+
|
Companies:
|
University of Arkansas for Medical Sciences
|
Address:
|
Arkansas Center for Birth Defects Research, Little Rock, AR, 72202, USA
|
Keywords:
|
microarrays ;
genomics ;
methylation ;
resampling ;
permuation test
|
Abstract:
|
In microarray experiments, chip-to-chip measurement bias can be a potential source of false associations if case/control status and other key covariates are not properly balanced or randomized in the experimental design. This has not always been widely appreciated, nor always mentioned in the protocols of commercially produced microarrays. Thus, it is not uncommon for a statistician to be presented with a dataset in which chip effect is completely or partially confounded with case/control status.
To properly control type I error in an experiment using the Illumina HumanMethylation27 microarray (Chowdhury et al 2011), we implemented a resampling-based approach in which an empirical null distribution of the t-statistic was generated by randomizing case/control status. The specific randomization procedure makes the null hypothesis artificially true, but retains the imbalanced experimental design and therefore the chance for spurious associations due to chip effect. Thus, the empirical null has heavier tails than the theoretical t distribution, which reduces power but properly controls type I error. In simulations, we estimate the reduction in power over a range of parameters.
|
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 2011 program
|
2011 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.