Dear All
Prof Xihong Lin of Harvard University of USA, the previous Co-Editor
of Biometrics and the COPSS President Award winner of 2006, will
give a seminar at School of Mathematics, The University of
Manchester on Weds 11th October 2006. The details are
Time & Date: 13:00 - 13:50, Weds 11th October 2006
Place: Room C.18, Ferranti Building, Sackville Street,
Manchester University
The title and abstract of the talk are attached below. You are
kindly reminded that following this seminar there is a Biostatistics
and RSS Local Manchester joint seminar at Manchester Dental
Education Centre (MANDEC), commencing at 14:00 on the same day.
Directions: The Ferranti building is No. 20 on the University
Campus map at
http://www.maths.man.ac.uk/mims/graphics/campus-map.pdf
----------------------------------------------------------
Title: Semiparametric Regression of Multi-Dimensional Genetic
Pathway Data: Kernel Machines and Mixed Models
Xihong Lin, Professor of Biostatistics
Department of Biostatistics
Harvard School of Public Health
We consider a semiparametric regression
model that relates a normal outcome to covariates and a genetic
pathway, where the covariate effects are modeled
parametrically and the pathway effect of multiple gene expressions
is modeled nonparametrically using least squares kernel machines
(LSKMs). The nonparametric function of a genetic pathway allows
for the possibility that each gene expression effect might be
nonlinear and the genes within the same pathway are likely to
interact with each other in a complicated way. This semiparametric
model also makes it possible to test for the genetic pathway
effect in a systematic way. We show that the least squares
kernel machine can be formulated using a linear mixed model.
Estimation and inference hence can proceed within the linear mixed
model framework using standard mixed model software. Both the
regression coefficients of the covariate effects and the least
squares kernel machine estimator of the nonparametric function
can be obtained using the Best Linear Unbiased Predictor (BLUP) in
the corresponding linear mixed model formulation. The smoothing
parameter and the kernel parameter can be estimated as variance
components using Restricted Maximum Likelihood (REML) in the
linear mixed model formulation. A score test is developed to test
for the genetic pathway effect. Model/variable selection within the
least square kernel machine framework is discussed.The methods
are illustrated using a prostate cancer data set and evaluated
using simulations.
|