INFORMATIVE GENE SELECTION USING BAYESIAN MODEL AVERAGING FOR PATIENTS SURVIVAL ANALYSIS

Choon, Yee Wen (2010) INFORMATIVE GENE SELECTION USING BAYESIAN MODEL AVERAGING FOR PATIENTS SURVIVAL ANALYSIS. Other thesis, Universiti Teknologi Malaysia.

[img] Microsoft Word
Choon_Yee_Wen.doc

Download (45Kb)

Abstract

Microarray technology is now widely used to identify potential biomarkers for cancer prognostics and diagnostics. With a vast number of genes information produced by microarray, informative gene selection is needed to both decrease clinical costs and mitigates the possibility of overfitting due to high inter- variable correlations. A problem with feature selection algorithms used to produce continuous predictors of patient survival is that they fail to explain the model (a set of selected genes whose regression coefficients haven been calculated for use in predicting survival prognosis) uncertainty. With thousands of genes and only tens to hundreds of samples, it often happens that a number of different models describe the data about equally well. In this research. BMA (Bayesian Model Averaging) method is applied to select a subset of genes for survival analysis on microarray data. BMA combines the effectiveness of multiple models by taking the weighted average posterior distribution instead of choosing a single model and proceeding as if the data were generated from it. In this research, BMA method showed that it can successfully select the related genes and produce significant result for the experiments in shorter time. The results obtained from BMA method proved that BMA is an appropriate method to use in gene selection.

Item Type: Thesis (Other)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Computer Science and Information System > Software Engineering
Depositing User: Unnamed user with email knizam@utm.my
Date Deposited: 04 Jul 2013 07:05
Last Modified: 04 Jul 2013 07:05
URI: http://ir.fsksm.utm.my/id/eprint/604

Actions (login required)

View Item View Item