Graphical models for describing relations between gene expression and schizophrenia

Specialeforsvar ved Michael Carter Bisgaard Galanakis 

Titel: Graphical models for describing relations between gene expression and schizophrenia 

Abstract: This thesis examines how graphical models can be used to analyse gene expression data. It focuses on solving the challenges that arise from handling data with degenerate zeroes, and implementing solutions that can handle the large data sets that result from RNA sequencing. The thesis uses data from the Lieber Institute for Brain Development to examine the gene expression of subjects with schizophrenia. The methods that were developed proved to be useful for finding genes with differential expression, and interesting structures of conditional independence between genes. However, how to interpret these structures remains a question for further research.

Vejledere:  Steffen Lauritzen, Lars Arvastson, Lundbech
Censor:       Søren Andersen, Novo