Abstract Details
Activity Number:
|
248
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 10, 2015 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract #317590
|
View Presentation
|
Title:
|
Prioritizing Hypothesis Tests for High-Throughput Data
|
Author(s):
|
Paul Schliekelman* and Sangjin Kim
|
Companies:
|
University of Georgia and University of Georgia
|
Keywords:
|
multiple testing ;
high throughput data ;
FDR ;
GWAS ;
gene mapping ;
RNA-seq
|
Abstract:
|
The massive increase in the resolution of biological data comes at the cost of increased multiple testing correction. As a result, the power for individual hypothesis tests often becomes very low. One solution is to use independent sources of information to identify the features of the data most likely to be true effects. Such filtering techniques have been in common use throughout the genomic era, but little is known about their statistical properties or the conditions which make them successful. We introduce a framework for quantifying the effectiveness of filtering strategies and show that a filter with a high probability of ranking true positives features highly (e.g. top 10%) can increase discovery probabilities for low effect size features dramatically, but filter effectiveness is highly dependent on the choice of filter cutoff point. We introduce a data-based method for choosing the filter cutoff point in a statistically rigorous fashion and show that its use can increase discovery probabilities by several-fold relative to a naïve choice of filter cutoff. We show various other results relating the filter properties to its effectiveness in increasing discovery probabilities.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2015 program
|
For program information, contact the JSM Registration Department or phone (888) 231-3473.
For Professional Development information, contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
2015 JSM Online Program Home
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.