Making Gene Sets More Coherent

  • Author / Creator
    Mahdavifard, Fariba
  • One important goal in microarray data analysis is to learn a predictor using a patient’s microarray data to predict some important characteristics of that patient. The high dimensionality of data makes learning such classifiers very challenging. We tried to use prior biological knowledge to tackle the challenges. Our colleagues have produced clusters of genes with a common function, called “PBT”s, for mouse and human. We hoped we could use each cluster as a single feature. This is most effective if each PBT is “coherent”. They expect all PBTs to be coherent; but while mouse PBTs are coherent, human PBTs are not. In this thesis we propose a method, called MkCoh, to improve the coherency of each PBT by removing and flipping some genes. We expected the predictors based on the revised PBTs to be more accurate than the ones based on either the original PBTs, or on the original gene expression values. However, our experimental results did not demonstrate this; we explored some possible reasons.

  • Subjects / Keywords
  • Graduation date
  • Type of Item
  • Degree
    Master of Science
  • DOI
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.