Event Date:
Event Date Details:
Wednesday November 6, 2024
Event Location:
- Zoom
Event Price:
FREE
Event Contact:
Saumyadipta Pyne, Ph.D
Related Link:
- Department Seminar
Abstract:
The talk will focus on our work on rigorous characterization of complex biomedical phenomena in healthy and disease conditions that are observed with high-dimensional data. Towards this, we develop new and precise representations of the human phenome, e.g., as curves and surfaces, that lead to accurate identification of various phenotypes including rare ones. Further, these representations allow for their optimal storage and search in databases as well as systematic comparison and clustering. Finally, we use finite mixture models for characterizing the phenotypic heterogeneity of the studied population with the help of clinical covariates. We will illustrate our approach with studies of such phenomena as neurodegenerative progression and intratumor heterogeneity.
Biosketch:
Dr. Saumyadipta Pyne's areas of research interest are computational statistics, machine learning and health data science. He received his Ph.D. in Computational Biology from the State University of New York at Stony Brook and trained as a postdoctoral associate at the Broad Institute of MIT and Harvard University. He held different positions at premier institutions such as MIT, University of Pittsburgh, Harvard Medical School, Indian Statistical Institute, Indian Institute of Public Health, and National Institute of Medical Statistics, New Delhi. He has served in many capacities including the PC Mahalanobis Chair, Full Professor and Head, Research Scientist, Scientific Director, Ramalingaswami Fellow, Senior Research Fellow and Visiting Professor. Dr. Pyne was a National Service DATA Scholar at NIH while being an Adjunct Professor at the Department of Statistics and Applied Probability at University of California Santa Barbara. He leads a team of data scientists and modelers at the Health Analytics Network.