Machine Learning in Medicine (MLxMed) Seminar Series

Speaker: Lana Garmire, PhD, Associate Professor, Computational Medicine and Bioinformatics, University of Michigan

Talk Title: Single-cell data science applied for cancer treatment and prognosis prediction

Talk Abstract: Recently years we have seen a wealthy collection of computational methods developed to address challenges in single cell data analysis. However, an essential remaining question is how clinical domains (eg. oncology) can benefit from it? In this talk, I will debrief some recent relevant endeavors in my research group. First, I will describe a new drug recommendation method called ASGARD, which uses the patient scRNA-Seq data to repurpose drugs at the personalized level, exemplified by breast cancer, leukemia and COVID cases. Next, I will go over new discoveries on a large population cohort of single-cell imaging mass cytometry data from breast cancer patients. We computed hundreds of tumor and tumor microenvironment level features, applied them to a deep-learning framework (Cox-nnet) developed by my group, and revealed novel breast cancer survival subtypes with atypic prognosis outcomes. Together these research projects highlight the promise of using single cell data and bioinformatics to guide personalized therapeutic treatment and to predict patient prognosis more precisely.

Location: Virtual Only.

Zoom Link: https://pitt.zoom.us/j/94397475670

If you have issues with the link above, please email Maria Ann Bond at bond@pitt.edu

For attendance details and more information, please visit: http://ml-in-medicine.org/