Abstracts Biological Sciences

A Translational Bioinformatics Approach to Parsing andMapping ISCN Karyotypes: A Computational Cytogenetic Analysis ofChronic Lymphocytic Leukemia (CLL)

by Zachary Abrams

Translational Bioinformatics is the field of study pertaining to the interpretation, analysis, and storage of large volumes of biomedical data for the purpose of improving human health. This thesis takes a translational bioinformatics approach through the large-scale analysis of karyotype data. Karyotyping, the practice of visually examining and recording chromosomal abnormalities, is commonly used to diagnose and treat disease. Karyotypes are written in a special language known as the International System for Human Cytogenetic Nomenclature (ISCN). Analyzing these karyotypes is currently done in a manual, non-computational manner due to the structure of the ISCN. The ISCN is generally considered not computationally tractable and as such precludes the potential of these genomic data from being fully realized. In response, this thesis presents the development of a cytogenetic platform (the Loss-Gain-Fusion model) that allows the transformation of human-readable ISCN karyotypes into a machine-readable model for computational analysis. This platform then utilizes text based cytogenetic data to create a structured binary karyotype language.Based on this computer readable language, several analyses are performed to demonstrate the potential of these data. First, the LGF model was applied to the Mitelman database (a publically-available karyotype database) to distinguish different diseases; in the process, we discerned which algorithms performed the best on the LGF data format. Second, an analysis was conducted to find potentially missed cytogenetic aberrations that recur in chronic lymphocytic leukemia from clinical data at the Ohio State university. Third, triplets containing drug, gene, and disease information were generated via a computational pipeline that connected various public drug-gene interaction data sources to identify potential drug repurposing hypotheses. The research presented here has detailed a novel approach to analyzing cytogenetic data. Advisors/Committee Members: Payne, Philip (Advisor).