Jenelle Bray, PhD

Photograph of Jenelle Bray

Distinguished Postdoctoral Fellow

Stanford University

Computational Structural Biology

jenelle@stanford.edu

Overview

I am a computational biologist, and my research focuses on the development and application of computational methods to understand and predict the structure, function and dynamics of proteins. I am currently a postdoctoral research fellow at Stanford University, working with Russ Altman and Michael Levitt in Simbios, the National Center for Physics-based Simulation of Biological Structures. As a postdoc, I work on two main projects: torsion angle normal mode analysis and using machine learning to discover novel protein ATP binding sites.

I got my Ph.D. in 2009 from California Institute of Technology (Caltech) in chemistry. I worked on GPCR structure prediction methods in Bill Goddard's group as a graduate student.

I have training from the Stanford Graduate School of Business in innovation in entrepreneurship through the Stanford Ignite program.

I am originally from Oregon, where I earned by B.S. in chemistry and mathematics at the University of Oregon, in the Clark Honors College. I developed my love of the outdoors growing up there, and I still believe Oregon has the best waterfalls, berries, and beer in the world.

Research

Torsion Angle Normal Mode Analysis

Normal mode analysis identifies the natural, resonant motion of a protein. It can be used to quickly model the global conformational changes of large proteins, when other methods like molecular dynamics are too computationally intensive to reach such long time scales. Traditionally, normal mode analysis has been carried out in Cartesian coordinates, but there are many advantages to calculating normal modes in torsion (dihedral) angle space. I work on developing torsion angle normal mode analysis methods, and I have shown that the method is faster, takes fewer degrees of freedom, and is more accurate at modeling protein conformational changes than Cartesian normal mode analysis. A new faster, more accurate normal mode analysis method is an important new tool for computational biologists because in addition to predicting conformational changes of proteins, normal modes can be used to help solve x-ray crystallography structures or to improve protein-ligand docking calculations for rational drug design.

For more detail on this research, read the technical paper.

Here's a short video of me giving an overview of this research.

For more information on the basics of normal mode analysis, read this article I wrote for the Biomedical Computation Review.

ATP Binding Site Recognition Through Machine Learning

ATP transports chemical energy within cells, and it is one of the most abundant molecules in cells. However, there is currently no method to identify ATP protein binding sites in proteins. I am working on using machine learning within FEATURE to make an ATP binding model to help experimental and computational biologists discover novel protein ATP binding sites.

GPCR Structure Prediction

G-protein coupled receptors (GPCRs) are a superfamily of transmembrane proteins that regulate signal transduction. They are targets of 50% of recently released drugs, and 25 of the top 100 best-selling drugs. However, the structures of GPCRs are very hard to determine experimentally, so there are very few solved 3D structures. Therefore, I developed computational methods to predict the structures and ligand binding of GPCRs, which are expected to aid in designing drugs that target GPCRs.

For more information about this research, read these technical papers: Paper 1 Paper 2 Paper 3, or my doctoral thesis.

Publications