Posted By: HGG Advances
Each month, the editors of Human Genetics and Genomics Advances (HGG Advances) interview researchers who have published work in the journal. This month, we check in with Rachel Sharp (Bluesky: @rrsharp.bsky.social) to discuss her paper “Considering social risk alongside genetic risk for bipolar disorder in the All of Us Research Program.”
HGGA: What motivated you to start working on this project?
RS: Questions surrounding psychiatric and community health have always been close to my heart. Before graduate school, I worked on an emergency room behavioral assessment team and volunteered as a crisis counselor, which helped me understand that it is impossible to separate a person’s psychological symptoms from their daily life. Most of the time, these illnesses don’t just appear out of nowhere; their onset and severity are deeply entangled with a person’s environment, financial stability, and support systems. When I moved into statistical genetics, I wanted to help bridge the gap between how we model polygenic risk and how psychiatric disorders actually manifest in real communities. For bipolar disorder, current polygenic risk scores explain only a small fraction of disease liability in curated research cohorts, and their predictive power drops even further in electronic health record biobanks and non-European populations. While geneticists widely recognize how important social determinants of health are, actually building them into our models is a major challenge. Doing this well requires cohorts like the All of Us Research Program, which was intentionally designed to capture social and environmental data at scale. I was thrilled at the opportunity to use this dataset to carefully measure and evaluate these lived experiences as the core drivers of risk we know them to be.
HGGA: What about this paper/project most excites you?
RS: The thing that excites me most about this project is how interdisciplinary it is by nature. My personal scientific (and life) philosophy is that no question has one answer. This project required me to think about neuropsychiatric risk from the perspective of a geneticist, a clinician, an epidemiologist, a statistician, a bioethicist, and critically, a patient. Each of those lenses prompted different questions about how we define phenotypes, handle missing survey data, and interpret risk, which ultimately led to much more nuanced answers. Because of the scope of this work, I got to collaborate with and learn from an amazing group of scientists across disciplines. It pushed me out of my comfort zone, and I believe both the paper and my perspective as a scientist are stronger for it.
HGGA: What do you hope the impact of this work will be for the human genetics community?
RS: I hope that this project, and the way we chose to approach the question, helps promote a broader, more holistic way of approaching human genetics. Studying human cohorts is inherently messy because there is never a true control group. Unlike studying model organisms in a lab where environments can be tightly regulated, human lives can’t be standardized, and it’s incredibly difficult to accurately disentangle genetic associations from environmental influences. However, as I think my paper and others in the space are starting to show, we can actually better parse disease liability when we directly account for the lived experiences that differentially shape our health. In our cohort, individuals who had the lowest genetic risk, but who experienced high social-risk burden, had bipolar disorder prevalences comparable to or exceeding those of people with the highest genetic risk and low social burden. These social factors reflect the everyday stressors and healthcare barriers that directly influence both mental health and access to care. Even though addressing them is complex, they highlight critical areas where community support, clinical resources, and policy changes can make a difference.
HGGA: What are some of the biggest challenges you’ve faced as a young scientist?
RS: One of the biggest challenges I’ve faced has been translating my foundation in psychology and neuroscience into the quantitative world of statistical genetics, without losing sight of the actual people behind my data. Moving from patient-facing roles into analyzing de-identified biobank data requires a deliberate shift in perspective. When you spend every day looking at diagnostic billing codes and survey scores on a computer screen, you have to be intentional about centering the human experiences that shaped those records. You need confidence in your data points, but I also know that a diagnostic code could reflect clinical reality for one patient and clinician bias for the next, while the absence of a code might simply mean a patient couldn’t afford to seek care in the first place. Coming from outside traditional genetics, I initially had to work hard to bridge those distinct scientific approaches. But I’ve come to realize that this outside perspective is actually my biggest strength. My multidisciplinary training helps me make connections across fields, question standard methodologies, and keep my research grounded in the communities it’s meant to serve.
HGGA: And for fun, what is one of the most fascinating things in genetics you’ve learned about in the past year or so?
RS: I recently read a fascinating study conducted in partnership with the Turkana people of Kenya that really challenges how we think about “risk” variants. While traditional genetics tries to isolate DNA by holding the environment constant, Indigenous scholars and community-partnered researchers have really pushed the field to honor environmental and cultural context. In this paper, researchers found variants in STC1 that are associated with markers of cardiometabolic risk, but only among Turkana people living in cities. In their ancestral desert environment, where pastoralists herd livestock under extreme water scarcity, those same variants actually protect kidney and metabolic health. If we only looked at urban clinical data, as most of our current genetic studies do, we might mislabel a protective adaptation as disease-causing. It’s a great reminder that for complex traits, genetic variants rarely have a single fixed effect because their impact on our health depends on the world we are living in.
Rachel Sharp, PhD, recently defended her thesis in the Neuroscience Curriculum at the University of North Carolina at Chapel Hill.