Elijah Roberts

Elijah Roberts

Assistant Professor

PhD, University of Illinois at Urbana-Champaign

Mergenthaler Hall 121 B
Group/Lab Website

Elijah Roberts is an assistant professor in the Department of Biophysics. His research focuses on investigating the creation and use of in silico cell models.

My laboratory is devoted to understanding and modeling the behavior of cells as complex systems. We are using tools from the general area of biological physics: potential- and probability-based computational modeling along with limited applications of single-cell, single-molecule experimental techniques (split roughly 80% theoretical/computational and 20% experimental). We term this approach “Physical Systems Biology” and it lies at the interface of biology, computer science, and physics.

While this approach absolutely requires in-depth characterization of particular components, an equally critical step is then stepping back to consolidate the knowledge gained into a model of the entire cellular system. Incorporating many varied types of biological data into a genuine in silico model of the cell is the long-range goal of the laboratory. We currently have three directions of research.

1. Spatial stochastic modeling

The laboratory's physical model of choice for studying stochastic, cell-scale processes on the time scales of cellular dynamics -- from one-to-several cell cycles -- is the reaction-diffusion master equation (RDME). RDME models are able to integrate a variety of cellular and biochemical data, e.g., from in vivo single-molecule, single-cell fluorescence experiments and cryoelectron tomography. We are developing new methods to improve the ability of RDME-based simulations to sample rare events (such as switching and decision-making processes). Many developmental processes involve a spatially determined switching and we would like to be able to spatially model the factors that control the switching decision. Our model system is the mating response of the yeast Saccharomyces cerevisiae.

2. Studying the physics of in vivo molecular systems

In order for probability-based RDME simulations, which can reach to cellular timescales, to benefit from studies of microscopic dynamics using potential-based simulations (molecular, Brownian, or Stokesian dynamics), which can reach only to micro-to-milliseconds, it will be necessary to develop a multiscale approach for spatial stochastic simulation of cells. We are studying ways to use potential-based simulations to parameterize local non-specific interactions for use in RDME. These parameters will be used to modify the diffusion and transition probabilities of mobile particles such that they account for the local environment of the cytoplasm.

Another limitation of the RDME, specifically in applying it to eukaryotic cells, is its dependence on diffusive transport of particles. Transport of many molecules in eukaryotic cells is not a diffusive process, but rather an active one. We are also studying multiscale approaches for incorporating force-based models of cytoskeletal transport into reaction-diffusion simulations.

3. Integrating stochastic and genome-scale models of cells

Transcriptional and translational regulatory networks control the phenotype of modern cells, regulating gene expression in response to environmental conditions and/or biological stimuli. Genome-scale models of cellular biochemical networks attempt to provide a static (steady-state) description of these phenotypes directly from the genotype. For the near-to-mid future it is unlikely that enough data will be available to construct a full kinetic model of the cell, including the regulatory networks. Therefore, we are working on approaches to gradually build stochastic dynamics into genome-scale models.

  • Elijah Roberts, John E Stone, Zaida Luthey-Schulten
    Lattice Microbes: high-performance stochastic simulation method for the reaction-diffusion master equation
    2012, Submitted
  • Michael Assaf, Elijah Roberts, and Zaida Luthey-Schulten
    Determining the stability of genetic switches: explicitly accounting for mRNA noise
    Phys. Rev. Lett., 106(24):248102 (2011)
  • Elijah Roberts, Andrew Magis, Julio O. Ortiz, Wolfgang Baumeister, and Zaida Luthey-Schulten
    Noise contributions in an inducible genetic switch: A whole-cell simulation study
    PLoS Comput. Biol., 7:e1002010 (2011)
  • Elijah Roberts, John E Stone, Leonardo Sepulveda, Wen-Mei W Hwu, and Zaida Luthey-Schulten
    Long time-scale simulations of in vivo diffusion using GPU hardware
    In Proceedings of the 2009 IEEE International Symposium on Parallel \& Distributed Processing (2009)
  • Elijah Roberts, Anurag Sethi, Jonathan Montoya, Carl R Woese, and Zaida Luthey-Schulten
    Molecular signatures of ribosomal evolution
    Proc. Natl. Acad. Sci. USA, 105(37):13953--13958 (2008)
  • Elijah Roberts, John Eargle, Dan Wright, and Zaida Luthey-Schulten
    MultiSeq: unifying sequence and structure data for evolutionary analysis
    BMC Bioinformatics, 7:382 (2006)