ML/Google Distinguished Lecture Series-Machine Learning Department - Carnegie Mellon University

Machine Learning Special Seminars


Spring 2017 Seminar
Date: April 11, 2017
Time: 4:30 PM - 5:30 PM
Location: 6115 Gates and Hillman Centers
Speaker: Anna Goldenberg
Title: Striving towards precision medicine: the machine learning approach
Abstract: Rapidly evolving technologies are making it progressively easier to collect multiple and diverse genome-scale datasets to address clinical and biological questions. How do we take advantage of this extensive and heterogeneous data to help patients? In this talk I will introduce several very different clinical questions that call for different machine learning approaches. The first question I’ll address is patient subtyping, i.e. identifying homogeneous subgroups of patients with the same diagnosis. I will introduce a network integration approach based on graph diffusion to help identify novel subtypes of disease. The second question I will address is identifying mechanisms of complex human diseases. To this extent we have developed a biologically inspired graphical model where we integrate multiple types of data and examine the posterior with respect to which genes of the mechanism are implicated in specific patients. And finally, if time permits, I will talk about our work on using deep generative models to predict drug response.
Speaker Bio: Dr Goldenberg is a Scientist in Genetics and Genome Biology program at the SickKids Research Institute and an Assistant Professor in the Department of Computer Science at the University of Toronto. She is also the fellow of the Canadian Institute for Advanced Research. Dr Goldenberg trained in machine learning at Carnegie Mellon University, with a post-doctoral focus in computational biology at UPenn and UofT. Dr Goldenberg’s lab develops machine learning approaches for networks and data integration using biological and clinical data. The current focus of her lab is on developing methods that capture heterogeneity and identify disease mechanisms in complex human diseases. Her translational focus is on methods that efficiently combine many types of patient measurements to refine diagnosis, improve prognosis and personalise drug response prediction for cancer patients.