Inferring and analyzing gene functional networks (graphs) from diverse and high-dimensional datasets - FLS Talk by Sara Mostafavi, UBC Statistics & Medical Genetics

Hugh Dempster Building (6245 Agronomy Rd.), Room 110

Speaker:  Sara Mostafavi, Assistant Professor of UBC Statistics & Medical Genetics and Associate Member of UBC Computer Science

Title:  Inferring and Analyzing Gene Functional Networks (Graphs) from Diverse and High-dimensional Datasets

Abstract:

Networks (graphs) offer a natural representation of complex systems, and as such they are ubiquitous in molecular biology. The recent availability of diverse and large-scale genomics data is enabling computational inference of large networks that summarize the relationships between genes and/or proteins, providing novel insights into a) gene function, and ii) disruption of regulatory relationships in disease. In this talk, I will present computational and machine learning approaches for inferring and deriving insights from gene functional networks. I will first describe a scalable method for the graph-based integration of diverse types of genomics data, in order to accurately infer functional roles for uncharacterized genes based on a small set of known (training) genes.  This approach results in the state of the art for automatically leveraging the continuous production of new genomics data. Second,  I'll describe an approach, based on sparse Gaussian Graphical Models, for joint inference of multiple related graphs. Applying this approach to 35 human tissues results in novel insights about tissue-specificity of regulatory relationships between genes.

Bio:

I arrived at UBC in January 2015, after a stop at the Benoist & Mathis lab at Harvard medical school in 2014. Before then, I was a postdoc at Stanford, working with Daphne Koller (now at Coursera). I got my PhD in Computer Science from the University of Toronto in 2011, where I worked with Quaid Morris. My thesis was on integrating large-scale genomics and proteomics datasets to predict gene function, check out GeneMANIA to find out more! You can find my CV here and my google scholar page here.