This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 407
Type: Contributed
Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract - #307160
Title: Using False Discovery Rates to Determine Cutoffs for Cluster Membership
Author(s): Johanna Hardin*+
Companies: Pomona College
Address: 610 N. College Ave., Claremont, CA, 91711,
Keywords: robust ; cluster ; biweight ; correlation ; microarray
Abstract:

Due to multiple comparison problems inherent in hypothesis testing for high-throughput data, using the false discovery rate (FDR) has emerged as the accepted tool for decision making. Within clustering there are no hypotheses to test, and so using FDR is not yet a standard aspect of most clustering algorithms. However, we argue that FDR is exactly the next step in making clustering algorithms biologically meaningful. For our model, we consider an observation to be "significant" if it belongs to at least one cluster, and the observation is "null" if it is not highly correlated enough to belong to any cluster. A test statistics (e.g., the biweight correlation as a robust measure of similarity) can be ranked, and a cutoff for significance is given based on the number (or rate) of acceptable false positives. We allow for observations to be allocated to no, one, or multiple cluster(s).


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 2010 program




2010 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.