The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
131
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 30, 2012 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Biometrics Section
|
Abstract - #306654 |
Title:
|
Principal Variable-Based Gene Selection for Expression Assay
|
Author(s):
|
Jie Ding*+ and Giovanni Parmigiani
|
Companies:
|
Dana-Farber Cancer Institute and Harvard School of Public Health
|
Address:
|
10 Peabody Terrace Apt 21, Cambridge, MA, 02138, United States
|
Keywords:
|
Gene expression ;
Gene selection
|
Abstract:
|
Although whole-genome gene expression essays are readily available nowadays, under certain situations it is still desirable to measure only a subset of genes, often with higher accuracy and reliability or in a tissue-specific design. We are investigating the problem of selecting a near-optimal subset of genes based on existing gene expression data in the absence of phenotypic information for the design of later assays. The goal is to identify a group of genes that best represent the variation of all the measured genes. This is analogous to the principal component analysis, with the key difference being that the subset here consists of individual genes instead of linear combinations of genes. We base our method on the idea of principal variables and take two different computational approaches: a general stochastic search algorithm and a fast stepwise algorithm. We apply both algorithms to simulated data and several cancer genome data sets. The resulting subsets of genes are evaluated for their abilities to predict expression levels of genes outside the subsets as well as phenotypes in the data. Our method is also applicable to other similar large-scale feature selection problems.
|
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 2012 program
|
2012 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.