Rosemary Braun (Northwestern) - Using Gene Expression to Tell Time

Return to Full Calendar
Date(s):
May 8, 2019 at 1:00pm - 2:00pm
Location:
Harper Center (Booth), Rm. 219
Event Audience:
all
Rosemary Braun

Speaker: Rosemary Braun Assistant Professor, Northwestern University

Rosemary Braun (PhD, MPH) is an Assistant Professor in the Division of Biostatistics (Dept of Preventive Medicine, Feinberg School of Medicine) and the Department of Engineering Sciences and Applied Mathematics at Northwestern University. Dr Braun’s overarching research interests are in the development of mathematical and computational methods to elucidate how large-scale biological phenomena emerge from the complex interplay of thousands of microscopic interactions. To this end, her research group develops machine-learning methods for the statistical analysis of high-dimensional omics data; graph-thoretical approaches for modeling the behavior of regulatory networks; and dynamical simulations to study molecular interactions. Recent publications have appeared in PNAS, Nucleic Acids Research, and Bioinformatics. Dr Braun obtained her PhD in Physics in 2004 from the University of Illinois at Urbana Champaign, where she studied the statistical physics of living systems under Klaus Schulten. This was followed by an MPH (Concentration in Biostatistics) at Johns Hopkins in 2006. Prior to joining Northwestern, she was a Postdoctoral Fellow at the National Cancer Institute (2006-2011).

Abstract: Using Gene Expression to Tell Time

Determining the state of an individual’s internal physiological clock has important implications for precision medicine, from diagnosing neurological disorders to optimizing drug delivery. To be useful, such a test must be accurate, minimally burdensome to the patient, and robust to differences in patient protocols, sample collection, and assay technologies. In this talk I will present TimeSignature, a novel machine-learning algorithm to estimate circadian variables from gene expression in human blood. By making use of the high dimensionality of the gene expression measurements and exploiting the periodic nature of the circadian variables we wish to predict, TimeSignature can be applied to samples from disparate studies and yield highly accurate results despite systematic differences between the studies. This generalizability is unique amongst expression-based predictors and addresses a major challenge in the development of robust biomarker tests. This talk will detail the method, present several applications, and discuss our recent work to extend it.

Sponsor: Machine Learning Seminar Series

Type: talk