University of Houston

Department of Computer Science

In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy

Musodiq Bello

Will defend his dissertation

Learning-based Segmentation Framework
for Tissue Images Containing
Gene Expression Data

Abstract

The use of biological imaging for scientific discovery continues to grow. Thus, it has become critical to many research efforts to be able to efficiently utilize the additional spatial information being generated. Gene expression studies at the cellular level are necessary for investigating how genes control cell type identity, cell differentiation, and cell-cell signaling. Accurately characterizing the location of gene expression in relation to the underlying anatomical morphology results in a greater understanding of the role of genes. There is an unmet critical need for reliable and automated computational tools to perform such an association efficiently for the over 20,000 genes in the mammalian genome. Our long-term research goal is to provide an interactive online database of spatial and temporal gene expression patterns in the context of mouse brain anatomy to facilitate further research in functional genomics.

In this dissertation, we present a new automatic method for segmenting images of mouse brain tissue sections that contain gene expression data into distinct anatomical regions and subregions. This allows expression patterns to be quantified and compared across images as well as over time. Our contribution is a novel hybrid atlas that utilizes a statistical shape model based on a subdivision mesh, texture differentiation at region boundaries, and features of anatomical landmarks to delineate boundaries of anatomical regions in spite of the high variation in gene expression images. We envision this computational tool as part of a web-based database for gene expression analysis. Such databases will enable biologists to query and cluster genes based on similarity of expression patterns, which may lead to significant breakthroughs in understanding biological processes.

Faculty, students, and the general public are invited.