Date of Award

Spring 1-1-2019

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

First Advisor

Naomi P. Friedman

Second Advisor

Marie T. Banich

Third Advisor

Matthew C. Keller

Fourth Advisor

Soo H. Rhee

Fifth Advisor

Marissa A. Ehringer

Abstract

The conception of the Research Domain Criteria (RDoC) apotheosized integration across differing biostatistical domains in studying psychiatric dysfunction. For example, an RDoC aim is the integration of MRI imaging and genetics, which represent differing foundations for the biological etiology of psychiatric and cognitive states. Though seemingly distal, recent work in both fields suggests that psychological phenomena are not represented by simple a proiri hypothesized candidates; instead we require large-data statistical approaches to find pathways for detection and treatment. That is to say, as genetics moves towards more whole-genome approaches, imaging moves towards more whole-brain approaches, and these whole-system approaches will likely be useful for research and clinical practice in the future. Imaging genetics, the field of study at the intersection of these biostatical perspectives, is unequipped to integrate these distal whole-system practices at the current time, with the most popular current approaches in imaging genetics resorting back to the antiquated candidate gene methods that failed to find replicable results. The purpose of this dissertation is to offer a framework for (some) integration in these fields at the whole-brain whole-genome (WBWG) level. While not exhaustive, the research, procedures, scripts, and methods here will help navigate this translational space. Study 1 demonstrates an approach to high-resolution mapping using the classical twin design to find patterns of association across the human cortex at high resolution for a dimensional measure of depression and a related psychiatric behavior. We then integrate these mapping results with results from the whole-genome and whole-brain literature more broadly in a big data framework. Study 2 demonstrates the effectiveness of whole-brain models as phenotypes for genetic studies of intelligence and discusses some practices for the application of whole-brain phenotypes in genetic studies. Finally, Study 3 conducts a whole-genome association study of cEF and uses the patterns across the genome to implicate particular biological/neurological pathways for analysis. The final chapter returns to discuss each study and how it fits into the general Whole-Brain Whole-Genome framework we lay out in chapter 1.

Share

COinS