ARC 2021 Ziv Bar-Joseph Abstract and Biography
Abstract
Functional genomics based personal medicine
While most of the genomics based personal medicine work focuses on information encoded in the DNA, other types of dynamic genomics data can provide critical and often complementary information. Such information can include the levels of genes in tissues or cells as a disease progresses, changes in the microbiome over time, and the response of specific cell types to drugs. In this talk I will discuss the use of computational and machine learning methods that can integrate different types of dynamic information to model individual responses. Once such personal models are learned, the methods can be used to predict potential treatments based on the unique response of each individual. I will present specific examples in which we used these methods to study potential treatments for viral infections, for treating lung disease and for improving growth of pre-term babies.
Biography
Dr. Bar-Joseph is the FORE Systems Professor of Computer Science at the Machine Learning Department and the Computational Biology Department in the School of Computer Science at Carnegie Mellon University and is the recipient of the 2012 Overton Prize in computational biology. His primary research areas are computational Biology, Bioinformatics and Machine learning, focusing on the analysis and integration of static and temporal high throughput biological data for systems biology. Based primarily on methods from machine learning, his group develops computational solutions to problems ranging from experimental design to data analysis, pattern recognition and the reconstruction of dynamic biological networks. Bar-Joseph is currently leading a large center, part of the HuBMAP NIH consortium, focused on developing computational methods to help build a detailed 3D map of the human body. He has also worked on improving algorithms for distributed computational networks by relying on our increased understanding of how biological systems operate and what makes them robust and adaptable.