|
Activity Number:
|
104
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #306228 |
|
Title:
|
Logistic and Probit Regression Modeling of Proteomic Mass Spectra in a Case Control Study on Diagnosis for Colon Cancer
|
|
Author(s):
|
Bart Mertens*+
|
|
Companies:
|
Leiden University Medical Center
|
|
Address:
|
Department of Medical Statistics, Leiden, 2300 RC, The Netherlands
|
|
Keywords:
|
logistic regression ; probit regression ; Bayesian analysis ; mass spectrometry ; proteomic diagnosis ; birth-death process
|
|
Abstract:
|
We adapt logistic and probit regression models for the evaluation of diagnostic potential of mass spectroscopic data in proteomics case control studies. Instead of a direct attempt to model the observed case-control data as regression on peaks, we parameterize the predictor as a linear combination of Gaussian basis functions along the mass/charge axis. The location of these basis functions is treated as a random variable and must be estimated from the data. A fully Bayesian implementation is pursued, which treats the number of functional components as a random variable. Calculations are implemented through birth-death process modeling. We evaluate the models on data from a randomized blocked case-control designed experiment, which was carried out recently at Leiden University (LUMC). The experiment compares spectra of serum samples of 63 colon cancer patients with 50 controls.
|
