2003 | OriginalPaper | Chapter
POE: Statistical Methods for Qualitative Analysis of Gene Expression
Authors : Elizabeth S. Garrett, Giovanni Parmigiani
Published in: The Analysis of Gene Expression Data
Publisher: Springer New York
Included in: Professional Book Archive
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
In many gene expression studies, the goals include discovery of novel biological classes and identification of genes whose expression can reliably be associated with these classes. Here we present a statistical analysis approach to facilitate both of these goals. The key idea is to model gene expression using latent categories that can be interpreted as a gene being turned “on“ or “off“ compared to a baseline level of expression. This three-way categorization is used for defining a reference in the unsupervised setting, for removing noise prior to clustering, for defining molecular subclasses in a way that is portable across platforms, and for defining easily interpretable probability-based distance measures for visualization, mining, and clustering.