Theodore Morley: Researcher

About Me

I am a scientist and software developer who has over six years of experience in researching artificial intelligence, genetics, and healthcare processes. I've been working as a data scientist in the Ruderfer lab, at Vanderbilt University Medical Center in Nashville Tennessee.

I grew up on Elm Gate Farm, my family's farm in Vermont, digging around in old junkyards and building small projects outside often. I love hiking, theater, and cooking. I received my degree in computer science from McGill University in 2018.

Selected Research

Here are some of my research projects, along with links to view the associated papers:

• Nature Medicine: Phenotypic signatures in clinical data enable systematic identification of patients for genetic testing: We demonstrated that a predictive model could replicate the same process by which physicians generated clinical suspicion of rare genetic disease in pediatric patients. Our model had high predictive performance, and could lead to increased access to genetic testing for those patients who otherwise may not have access to appropriate care. The model from this paper is currently being evaluated for production use in the VUMC healthcare system.

  I'm proud to say our work was recognized as one of the top ten advances in genomic medicine in 2021 by the AJHG Year in Review.

• Preprint: Evaluating the impact of modeling choices on the performance of integrated genetic and clinical models: This paper is an investigation into how best to integrate genetic information (using polygenic risk scores) into existing predictive models which are based off of clinical data. We evaluated the point of data integration, complexity of the model, source of the data, amount of information available for patients, and more.

I've included only first author publications here, but Google scholar is usually quite good about collecting links to all the publications I've been a part of.

Presentations and Talks

• American Society of Human Genetics 2022: I gave the oral platform presentation Incorporating genetic information into clinical prediction models: Assessment of technical considerations and value in Crohn's disease as part of the session Polygenic prediction of complex disease.
• World Congress of Psychiatric Genetics 2021: Presented as part of the symposium "Discoveries from the Use of Polygenic Scores in an Electronic Health Record Research Setting", discussing early results in a project investigating methodology for combining genetic and phenotypic information in predicting Crohn's Disease.
• World Congress of Psychiatric Genetics 2019: Gave an oral platform presentation about ongoing research in prediction of genetic disease using electronic health record data, aiming to shorten the diagnostic odyssey of patients with rare disease.

Contact me

You can email me by taking the following address, replacing [at] with @, and "tree" with the tree type in the name of the farm I grew up on (mentioned at the start of this page):

I have my direct messages open on most websites that you can reach out to me on, and I'm happy to talk about science, software, or most things really.

Here's a picture of what I look like!