Activity Number:
|
304
- Statistical Learning: Dimension Reduction
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 1, 2017 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
Abstract #324366
|
View Presentation
|
Title:
|
Supervised Dimension Reduction with Application to Driver Gene Detection
|
Author(s):
|
Yichen Cheng*
|
Companies:
|
|
Keywords:
|
copy number variation ;
variable selection ;
L0 penalty
|
Abstract:
|
Copy number variation (CNV) refers to a copy number change that deviates from the normal states of 2 copies of DNA. The CNV is largely related with the tumor growth. So successful detection of CNV that is associated with tumor growth is an very important topic and the gene within such region is called the driver gene. The detection of driver gene can be difficult due to the high dimensionality of human genome and the fact that CNV may happen at different locations for different person. In this paper, we proposed a supervised dimension reduction technique that helps to identify the common pattern across subject. And the supervised learning ability helps us to identify patterns that are associated with the outcome. We developed an efficient algorithm for this purpose. We use simulation studies and real data examples to show the power of proposed method.
|
Authors who are presenting talks have a * after their name.