-----------------------------------------------------------
Statistics Seminar Series
Wednesday, Feb. 27, 2013, 2:30 PM
Cullimore, Room 611
New Jersey Institute of Technology
-----------------------------------------------------------
Integrative Analysis of
Prognosis Data on Multiple Cancer Subtypes
Using Compound Group Bridge
Jin Liu
Department of Biostatistics, Yale University
Abstract
In cancer research, profiling studies have been extensively
conducted, searching for genes/SNPs associated with prognosis. Examining the
similarity and difference in the genetic basis of multiple subtypes of the same
cancer can lead to better understanding of their connections and distinctions.
Integrative analysis approaches analyze the raw data on multiple subtypes simultaneously.
In this study, prognosis data on multiple subtypes of the same cancer are analyzed.
An AFT model is adopted to describe survival. The genetic basis of multiple
subtypes is described using the heterogeneity model, which allows a gene/SNP to
be associated with the prognosis of some subtypes but not the others. A
compound penalization approach is developed to conduct gene-level analysis and
identify genes that contain important SNPs associated with prognosis. The
proposed approach has an intuitive formulation and can be realized using an
iterative algorithm. Asymptotic properties are rigorously established.
Simulation shows that the proposed approach has satisfactory performance. An
NHL (non-Hodgkin lymphoma) prognosis study with SNP measurements is analyzed.