Informatics Insights with Theresa Walunas

Theresa Walunas, PhD, FAMIA, is an associate professor of Medicine in the Division of General Internal Medicine and Geriatrics, Microbiology-Immunology and Preventive Medicine in the Division of Health and Biomedical Informatics. Walunas leads a research laboratory at the intersection of biomedical informatics and implementation science, using machine learning and artificial intelligence for data integration, to identify complex inflammatory conditions in electronic health record data and develop population health management strategies. She has a strong commitment to graduate education and mentorship. She is director of the Master of Science in Health and Biomedical Informatics Program, Chair of the Health and Biomedical Informatics track for the Health Sciences Integrated PhD Program, and Associate Director for Graduate Studies for the Medical Scientist Training Program.

What’s your background?

I have a BS in Biology from Trinity University, a PhD in Immunology from the University of Chicago, an MS in Computer Science from DePaul University, and I have a lot of practical business experience from my time in a microbial genomics start-up company. I’m a boundary spanning biologist.

How did you get into the informatics space?

My standard joke is that I was the person in my PhD lab who understood the email system, which, by the standards at the time, made me the computer scientist.

But truthfully, my inspiration for informatics came from three places: first, my dad, who bought one of the first Apple II computers when I was 12 and gave me the chance to try my hand at telling computers what to do (which I loved!), next the Human Genome Project and the excitement about all the data that would come from it, and finally, an amazing molecular biologist in my post-doc lab who kept getting frustrated with the very biologist-unfriendly early informatics tools. I wanted to figure out how to make insights from data more accessible to my colleagues.

Social engineering is more important than software engineering. To move data, you need to move people first. Have a passion for using data to do good and make friends so that you can go further faster — and have more fun.”

What was some of your earlier work? What are you working on now?

My PhD research involved understanding the function of CTLA-4 — a protein on the surface of T lymphocytes. My work was the first to demonstrate that CTLA-4 was a shut off signal (now called a “checkpoint”) — this work was the foundation for modern immunotherapy for cancer. I had a break from immunology while I focused on microbial genomics and developing systems for microbial genome analysis. When I came to Northwestern it was to support the federal electronic health record (EHR) adoption program — and that opened my eyes to the possibility of modeling human immunologic disease with clinical data as well as trying to figure out how to improve care for immune disease patients based on that modeling. Right now, I’m working on developing surveillance strategies for immune-related adverse events following immunotherapy for cancer, supporting medical informatics infrastructure for two large NIAID funded programs (SCRIPT and NeuLung) focused on lung biology and immunology, and planning for the next phases of prior projects focused on developing identification and modeling strategies for lupus, cancer-associated cachexia, glioblastoma, and multisystem inflammatory syndrome in children.

What has been your greatest challenge?

Well, practically, time management. I want to learn all the things and there just aren’t enough hours in the day. But functionally, as someone who spans a lot of boundaries, it is learning to speak the languages of other researchers quickly so that I can find opportunities to collaborate that bridge my bench background and medical informatics expertise. My dream is to help develop more systems that really bring together pre-clinical research and molecular pathology data with EHR data to support high quality human models of immune disease.

Would you like to share a publication or project you are most proud of being associated with?

This is a tough one for a boundary spanner. I’ll give you one from each of my research “eras":

Walunas TL, Lenschow DJ, Bakker CY, Linsley PS, Freeman GJ, Green JM, Thompson CB, Bluestone JA. CTLA-4 can function as a negative regulator of T cell activation. Immunity. 1994 Aug;1(5):405-13. doi: 10.1016/1074-7613(94)90071-x. PMID: 7882171.

This is the true OG paper for me — it made a meaningful contribution to immunology, but it is also the paper where I really learned to trust my instincts and believe in my ability to do science.
Walunas TL, Ye J, Bannon J, Wang A, Kho AN, Smith JD, Soulakis N. Does coaching matter? Examining the impact of specific practice facilitation strategies on implementation of quality improvement interventions in the Healthy Hearts in the Heartland study. Implement Sci. 2021 Mar 31;16(1):33. doi: 10.1186/s13012-021-01100-8. PMID: 33789696; PMCID: PMC8011080.

Intersecting informatics and implementation science to understand what makes practice coaching tick. This paper was an incredibly fun collaboration and helped me understand the power of machine learning to understand patterns in data that could make real impact in clinical care settings.
Forrest N, Guggilla V, Bell A, Zelisko S, Federico EM, Power EA, Birch S, Nandoliya KR, Houskamp EJ, Tran S, Lukas RV, Johnson JL, Roy I, Wainwright DA, Walunas TL. Natural language processing algorithms identify wild-type isocitrate dehydrogenase gliomas in electronic health records. Neurooncol Adv. 2025 Jun 7;7(1):vdaf111. doi: 10.1093/noajnl/vdaf111. PMID: 40703809; PMCID: PMC12284639.

So, I don’t even know how to pick a phenotyping paper. I love all of them! But this one really puts everything together (including all my students at the time!) to develop a beautiful and practical computational phenotype. It touches on a variety of phenotyping strategies and has one of the most beautiful and elegant data validation approaches I’ve been a part of.

Informatics and data science are very collaborative. Who do you collaborate with?

Oh, this is a big list. I’ve been fortunate to collaborate with Abel Kho, MD, across my entire career at Northwestern. He’s opened a lot of doors for me to both people and data and a number of my papers and projects have been built from data from the CAPriCORN network he leads. I have also had amazing collaborations in the Lurie Cancer Center that have led to longer term collaborations with researchers at Shirley Ryan Ability Lab and Loyola University. My newest collaborations are with a great group of people in the Division of Pulmonary and Critical Care Medicine as part of the SCRIPT (NIAID Systems Biology Center) and Neu Lung (NIAID Human Immunology Center) programs. I’ve also had the good fortune to work with both the Illinois and Chicago Departments of Public Health. On a broader national scale, I’ve been involved in the eMERGE network, and several AHRQ-funded EvidenceNOW programs focused on quality improvement in pragmatic settings (H3, INSPIRE, HH4M) that involved organizations across the country.

What advice would you give to a student wanting to get into this field of study?

What are you excited about in the field of informatics right now?

Honestly, the people and the incredible student talent that I see coming into the field. They are teaching me new things all the time, including seeing things in different ways. On the technology front I’m really looking forward to seeing how large language models help us unlock information in text-based data from clinical teams and patients.