Joseph Romano, PhD, MPhil, MA
Joseph Romano, PhD, MPhil, MA
Joseph Romano, PhD, MPhil, MA is an Assistant Professor of Informatics in the Department of Biostatistics, Epidemiology and Informatics, Senior Fellow in the Institute for Biomedical Informatics (IBI) and an investigator in the Center of Excellence in Environmental Toxicology (CEET). His research leverages artificial intelligence (AI) and machine learning (ML) to enhance our understanding of how diverse environmental exposures, such as toxic chemicals, social determinants of health, or pharmaceutical drugs, result in human disease. With this aim, Romano Lab conducts original research at the interface of clinical/translational informatics and environmental health.
Prior to joining the DBEI in 2023, Dr. Romano earned a BS in Molecular Genetics from the University of Vermont (2014) and PhD, MPhil, and MA degrees in Biomedical Informatics from Columbia University, where he also completed a graduate research fellowship sponsored by the U.S. National Library of Medicine. After earning his PhD (2019), he completed a postdoctoral fellowship in environmental health science at the University of Pennsylvania.
In this Q&A, Dr. Romano shares what sparked his interest in biomedical informatics, how AI can support advancements in environmental health science, and what he enjoys most about being a part of the DBEI community.
Can you tell us about your current research and what inspired you to pursue this area?
In the broadest sense, my lab’s research is focused on using structured knowledge and artificial intelligence (AI) to predict: “If I’m exposed to this thing in the environment, what kinds of diseases might result from that exposure, and why?” Environmental exposures are very broad, including pharmaceutical drugs, toxic environmental chemicals, social determinants of health, and factors related to the ecosystem in which you live, among many others. We’re interested in all of them. For disease outcomes, we are especially interested in women’s health conditions and autoimmune-related side effects resulting from cancer immunotherapy. As for data science and informatics methods we use to study these phenomena, we are doing a lot of work in building biomedical knowledge graphs—complex network-like structures describing relationships between biomedical concepts—and using deep learning to find meaningful patterns in those knowledge graphs.
My interest in biomedical informatics started when I was in high school, when my biology teacher (Mrs. Raiford) saw that I had plans to study computer science, but she thought I could do well in biomedicine. She told me to look into this field that combines principles from both disciplines, and I was hooked. Since then, I’ve gravitated towards translational bioinformatics, which is a subfield that translates molecular observations into population health outcomes, and my amazing lab members have taught me about the disparities in women’s health and the great need for data-driven science to improve outcomes for these and related illnesses.
How does your work in Informatics intersect with real-world challenges?
The types of environmental exposures we’re interested in are things that regular people in the Philadelphia region (and globally) interact with every day. From chemicals in the water supply, to social determinants that are highly impactful in environmental justice communities around the city, and to our proximity to vehicle emissions from the I-95 corridor, these are exposure events that are constantly impacting our long- and short-term health. However, we know very little about what specifically most of these exposures do to the human body. If we were to try and study each one individually, there is no way we could tackle the millions of types of exposures with currently available tools. By using AI and data-driven techniques, we can at least prioritize the exposures that are most likely to be harmful and influence behavioral health and policy in ways that can have the greatest impact on the largest number of people.
What do you find most rewarding about working in the Department of Biostatistics, Epidemiology and Informatics?
DBEI’s greatest strength—aside from the many amazing faculty, staff, and trainees—is that it is truly an interdisciplinary department, bringing together three sub-communities that are usually isolated at other institutions. This is one of the key reasons why collaborative science is so accessible at Penn, and more specifically in the Perelman School of Medicine. We have strong ties to other centers and institutes, including the Center for Clinical Epidemiology and Biostatistics (CCEB), Institute for Biomedical Informatics (IBI), and others, which further strengthens the reach and impact of the work we do here. And since everyone in our community is committed to cutting-edge research in their own areas of research, the number of ways the DBEI community is impacting local, national, and global health is especially vast. Of course, none of this would be possible without the strong sense of community and welcoming culture that makes everyone so approachable and eager to work together to accomplish larger goals.
Can you share a significant recent project, publication, or professional recognition that you are particularly proud of contributing to or achieving?
As I mentioned earlier, we’re involved in several diverse areas of human health, but one that I’m especially proud of is a body of papers, grants, and ongoing research projects related to developing a new resource named “ComptoxAI.” This is the central focus of an NIH/NLM “Pathway to Independence Award” that has been supporting my lab since launching it in January 2023. We’re building ComptoxAI to be a major knowledge base in chemical and predictive toxicology, currently holding over 1.2 million chemicals of toxicological concerns and the diverse ways those chemicals are known to act on the human body. The main end-goal of this research is to provide a free and open-source knowledge base with built-in tools for environmental health scientists to build and run advanced AI models without needing to be experts in machine learning and data science. The environmental health science community is almost ubiquitously calling for the use of AI to advance prediction of how toxic exposures act on the body, but there are very few tools that are well-suited to support this work. As such, we’re thrilled that this body of research involving ComptoxAI has been very well received by the community and lays the groundwork for new and exciting research in the years and decades to come. We will be expanding ComptoxAI to include other aspects of the human exposome (lifestyle factors, social determinants, etc.) soon.
What advice would you give to students or early-career professionals in your field?
Biomedical informatics is a phenomenal area of study for trainees and other early-career scientists for many reasons (including career prospects), but our professional communities are a particular strength. Make sure to take advantage of these whenever possible! Depending on your area of research, get involved in AMIA/ISCB/ASHG or whichever societies are most relevant, and do it early. Attend conferences and present your work, and you will naturally form career-long connections with leaders in the field. Don’t be afraid to explore opportunities in both industry and academia, and remember that there are plenty of synergistic opportunities for both industry and academia to work together, especially in informatics. Finally, don’t short-change yourself on developing your coding skills! Even if you perform mainly qualitative research, computer programming and data literacy will impact every professional opportunity through your informatics career, and being well-equipped will take you far.
What is an interest or pastime that you enjoy outside of academia?
I have a few hobbies, but probably the most rewarding (and longest standing) is my involvement in music. I started learning piano when I was about 8 years old and picked up trumpet and a few other instruments in high school. In college, before switching to a science major, I attended music school and studied both classical piano and jazz trumpet and had the opportunity to play with some pretty awesome bands and musicians. Nowadays, I have a piano that I play virtually every day, and my wife and I are members of our church’s choir. We also attend a lot of concerts! Philadelphia is an amazing city for live music – we see a lot of shows ranging from classic and contemporary rock at the Mann Center, to orchestral concerts at the Kimmel Center, and jazz shows at Chris’ and other places around the city.