[PAST EVENT] Mathematics Colloquium - Dr. Mahlet Tadesse (Georgetown University)
Access & Features
- Open to the public
Title: Uncovering cluster structures and relevant biomarkers in “-omics” data
Abstract: High-throughput “-omics” technologies (genomics, epigenomics, transcriptomics, proteomics, metabolomics, etc) allow the simultaneous quantification of thousands of biomarkers. These technologies hold great potential for gaining insights into the complex biological processes underlying specific phenotypes and for identifying biomarkers that can be used for improved diagnosis and therapeutic interventions. The challenges of analyzing the generated data have led to the development of various statistical, computational and bioinformatic tools over the the past couple of decades. In this talk, I will present some of the methods we have proposed for uncovering cluster structures and relevant biomarkers by combining ideas of mixture models and variable selection. I will discuss (1) a bi-clustering approach that allows clustering on subsets of variables to refine disease classes and identify discriminating biomarkers, (2) an integrative model to relate data from different -omic levels using a stochastic partitioning method, and (3) a mixture of regression trees approach to uncover homogeneous disease subgroups and their associated predictors accounting for non-linear relationships and interaction effects. I will illustrate the methods on various -omic studies.
Organized by: WISE (Women in Science Establishment)