|
Activity Number:
|
196
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #308955 |
|
Title:
|
Model-Based Approach for Cancer Outlier Differential Gene Expression Detection
|
|
Author(s):
|
Baolin Wu*+
|
|
Companies:
|
The University of Minnesota
|
|
Address:
|
420 Delaware St SE, Minneapolis, MN, 55455,
|
|
Keywords:
|
Differential expression detection ; K-means clustering ; Least absolute deviation ; Microarray ; Outlier ; Partition around medoids
|
|
Abstract:
|
We study model based approaches to detect cancer genes that are over/down-expressed in some but not all samples in a disease group. This has proven quite useful in cancer studies since heterogeneous oncogene activation patterns have been observed in the majority of cancer types. In this paper we propose the model based outlier differential expression (MODE) detection methods, which are designed to automatically identify cancer outliers and quantify the differential expressions simultaneously. The proposed MODE can be derived from the likelihood ratio test under the Gaussian and exponential error models, and efficient numerical algorithms are developed for the maximum likelihood ratio computation. Using real and simulation studies, we compare the proposed MODE to the existing outlier differential expression detection methods in the literature and illustrate its competitive performance.
|
- 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 2007 program |
