Activity Number:
|
464
- Novel Approaches for Complex Biomedical Data
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
|
Sponsor:
|
Section on Statistics in Genomics and Genetics
|
Abstract #313517
|
|
Title:
|
A Modified Sam-T Test Statistic for Differential Expression
|
Author(s):
|
Sherwin Cedeno* and Isaac Dialsingh
|
Companies:
|
The University of the West Indies and The University of the West Indies
|
Keywords:
|
Differential Expression;
Significance Analysis of Microarrays;
Modified SAM;
Neuhäuser’s test;
proportion of null hypothesis;
Welsh's Test
|
Abstract:
|
Many statistical tests exist for the identification of genes that are differentially expressed. A variant of the SAM-t test statistic is derived and we compared this via simulations with existing tests. Our variant uses a cluster specific value of the tuning parameter rather than a universal tuning parameter for so in the SAM-t. Our test statistic’s performance is compared with competing tests such as Welsh’s and the Neuhäuser’s test under simulated data with varying configurations of pi0. We also devise a method for estimating pi0 based on the modified SAM-t test statistic. Our method is applied to a real dataset.
|
Authors who are presenting talks have a * after their name.