Penn State University
154 Hurley Hall
Detecting rare cells in single-cell data for HIV vaccine development
Current vaccine development for HIV has been targeted to induce protective T cells. Clinicians and immunologists rely on single-cell technologies to distinguish and identify functionally distinct T cell responses to vaccination. The ability to efficiently identify these cell subsets, especially the small ones is crucial to decipher system-level biological changes. During this talk, I will first present a new clustering framework names Hidden Markov Model on Variable Blocks (HMM-VB). HMM-VB leverages prior information about chain-like dependence among groups of variables to achieve dimension reduction as well as incisive modeling of the rare clusters. In the second part of the talk, I will briefly discuss a novel computational framework to assess the uncertainty inherent in clustering analysis, which is caused by randomness in data or limitations of algorithms.
Originally published at acms.nd.edu.