Document Server@UHasselt >
Education >
School for Information Technology >
Master theses >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/12526

Title: Using gene expression data to predict dose-response curves
Authors: Haldermans, Philippe
Advisors: SHKEDY, Ziv
Issue Date: 2010
Publisher: tUL Diepenbeek
Abstract: DNA microarrays allow us to monitor thousands of genes simultaneously. One area of interest is oncology, where we want to predict the response of a patient to a certain compound, using the microarray data. The burden of microarrays is the fact that the number of variables (p) is much larger than the number of subjects (n). In this thesis, we compare existing large scale prediction methods, such as Supervised Principal Component Analysis and Lasso, to our newly developed method called Weigted Ensemble Prediction. The method removes the uninformative genes before applying the prediction method. After a univariate screening of the genes using correlations, we select subsets of genes on which we apply the prediction method. After X times we collect all the fitted models and record which genes where used regularly in the models. These genes are then used for the final prediction step. We will show that this methods gives more accurate predictions compared to the existing methods.
Notes: Master of Statistics-Biostatistics
URI: http://hdl.handle.net/1942/12526
Category: T2
Type: Theses and Dissertations
Appears in Collections: Master theses

Files in This Item:

Description SizeFormat
N/A1.34 MBAdobe PDF

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.