Award Date

8-1-2019

Degree Type

Thesis

Degree Name

Master of Arts (MA)

Department

Psychology

First Committee Member

Daniel Allen

Second Committee Member

Mira Han

Third Committee Member

Andrew Freeman

Fourth Committee Member

Rochelle Hines

Fifth Committee Member

Kaushik Ghosh

Number of Pages

62

Abstract

Bipolar disorder (BP), a severe mental disorder characterized by recurring manic and depressive episodes, has been shown to have a strong genetic underpinning. Current theory suggests that it is the summation of risk alleles, spread across the entirety of the genome, which contributes to the development of BP, as well as other polygenic traits. The comorbid nature of these polygenic traits are often problematic for diagnosticians as the symptomology of the disorders may vary substantially between individuals and can create diagnostic confusion. To alleviate issues such as these, a more objective measure, to be used alongside current diagnostic procedures, is needed. To accomplish this, researchers have begun to turn their attention towards an ever increasing body of publicly available genetic data.

Recently, polygenic risk scores have been implemented in genetic risk prediction. Genome-wide association study (GWAS) summary statistics, derived on a plethora of psychiatric disorders, are readily accessible and provide a cost efficient strategy for generating risk scores. In this study, we attempted to not only predict the diagnosis of bipolar disorder utilizing publicly available genotype information, but to also improve upon current methodology by showing that the inclusion of risk scores calculated on comorbid traits can benefit the accuracy and generalizability of the classification model. While the results reported herein are mixed, this study provides strong support for the feasibility of genetic prediction of psychiatric disorders. This approach was, to our knowledge, entirely novel and the first time it had been implemented in practice.

Disciplines

Bioinformatics | Genetics | Psychology

Language

English


Share

COinS