Canine Cancer
Using spatial transcriptomics to understand mechanisms responsible for canine lymphoma progression
This project is part of the SDS 389 Research Methods course that is managed by the Center for Indisciplinary Research within the Statistics and Data Science program at St. Olaf.
Fall 2026 - Spring 2027
Industry collaboration with:
Dr. Shruthi Naik, Mayo Clinic
Elise Hachfeld, Mayo Clinic
Student Researchers:
Andi Huselid, Biology, Mathematics ’27
Lily Jacobsen, English, Mathematics ’27
McKelvey McDaniel, Statistics and Data Science, Mathematics ’27
Dogs and humans share environmental exposures and develop strikingly similar cancers, which makes naturally occurring tumors in pet dogs a powerful model for human disease. Lymphoma is a case in point: it is one of the most common canine malignancies, occurring several times more frequently in dogs than in people 1, yet the mechanisms driving its formation and progression remain poorly understood. Canine cancer data also carries a practical advantage for a data-focused researcher: because it raises few of the privacy concerns that tightly restrict human patient data, it can be shared openly. The NCI’s Integrated Canine Data Commons, for example, makes canine cancer data fully open-access and citable in publications 2. Despite this accessibility and the clear translational payoff comparatively few researchers work in this space, which leaves unusual room for motivated students to make a genuine contribution.
Investigators at Mayo Clinic are studying a novel immunotherapy for canine lymphoma and generating rich, high-dimensional datasets in the process. This project uses that data to ask two questions: (1) using spatial transcriptomics (pairing each cell’s gene expression with its physical location in the tissue), we aim to identify the cell of origin and characterize the tumor microenvironment, enabling direct comparison with human lymphoma; and (2) using gene-expression data from the immunotherapy trial 3, we aim to better understand the biological effects of immunotherapy on canine lymphoma in responding versus nonresponding patients.
Students working on this project will gain hands-on experience with high-dimensional spatial data analysis, supervised and unsupervised machine learning, and large-scale hypothesis testing. It’s a strong fit for self-directed, curious students who like connecting ideas and reading widely; prior math or machine learning exposure helps but isn’t required, and no biology background is needed.
Footnotes
AKC Canine Health Foundation. (2009, August 30). Canine lymphoma. https://www.akcchf.org/disease-history/canine-lymphoma/↩︎
National Cancer Institute. (n.d.). Integrated Canine Data Commons (ICDC). Cancer Research Data Commons, National Institutes of Health. Retrieved July 3, 2026, from https://datacommons.cancer.gov/repository/integrated-canine-data-commons↩︎
Anai LA, Cavalca AMB, Xavier PLP, Semolin LMS, Jark PC, Senhorello ILS, Fukumasu H, Munhoz TD, Figueiredo ML, Fernandes CC, Costa MT, Kobayashi PE, Santana ÁE, Fonseca-Alves CE. Transcriptomic profiling reveals prognostic gene signatures and immune landscape alterations in canine diffuse large B-cell lymphoma. Vet Oncol. 2026;3(1):1. doi:10.1186/s44356-025-00051-2. https://pmc.ncbi.nlm.nih.gov/articles/PMC8939155/, https://link.springer.com/article/10.1186/s44356-025-00051-2↩︎