Northwestern University

Tue 12:00 PM

BIDS Seminar: Sahil Shah - Student Presentation (Network-Based Approaches to Identify Disease Genes from Expression Data)

recurring see all events in this series

When: Tuesday, November 14, 2017
12:00 PM - 1:00 PM  

Where: Arthur Rubloff Building, Lakeview Conference Room (11th Floor), 750 N Lake Shore Dr, Chicago, IL 60611 map it

Audience: Faculty/Staff - Student - Post Docs/Docs - Graduate Students

Contact: Lindsay Varasteh   312.503.1997

Group: Center for Data Science and Informatics (CDSI)

Category: Academic


The ability to profile the expression levels of thousands of genes simultaneously and identify the genes associated with a disease has opened new avenues in understanding disease mechanisms and developing precision medicine interventions.  Since the organization of physical and functional cellular networks into databases, it has been possible to develop methods that analyze expression data in the context of these networks. A key challenge is to combine the expression data with the systems-level information and still obtain specific molecular targets. We present a new analysis technique, which we call GeneSurrounder, that identifies specific disease-associated genes and takes into account the complex network of cellular interactions. GeneSurrounder identifies genes that (i) appear to influence nearby genes on the network that (ii) themselves are dysregulated and associated with the disease under study. We apply GeneSurrounder to three distinct ovarian cancer studies using a global KEGG network and show that our method yields more consistent results across multiple studies of the same phenotype than competing methods. Because GeneSurrounder finds disease genes that match this specific signature in the data, yet can miss genes that may exhibit a different pattern in the data, we also present preliminary results for a new analysis technique that uses the same expression and network features as GeneSurrounder to search for a common pattern amongst the disease associated genomic variants (obtained from GWAS data) to then identify new genes that exhibit the same learned pattern. These methods can open up new avenues of precision medicine by identifying disease-associated genes.

Add Event to Calendar

Add Event To My Group:

Please sign-in