Theory of Coarse-Graining and Multiscale Phenomena
The Voth group develops theoretical and computational frameworks to study multiscale phenomena in biology and materials science. We work on both new theoretically-informed conceptual models as well as more quantitative, systematic, bottom-up modeling methods. Since 2005, our group has been one of the leaders in the development of methods for defining coarse-grained interactions from atomistic simulations in the condensed phase, and we have since contributed to systematic methods for defining optimal choices of coarse-graining strategy for large biomolecules and to conceptual CG models of membranes and fluids at the mesoscale.
Multiscale modeling is an important and growing field, but often, the connection between a coarse-grained model and the system of interest can be unclear and difficult to check if the model is not developed carefully. Our bottom-up approach has been set apart by its strong foundations in statistical mechanics and its quantifiable connections between fine and coarse models. However,while powerful, the bottom-up approach is expensive and limited in applicability, which has motivated us, in collaboration with the Center for Multiscale Theory and Simulation, to develop novel ways of thinking about coarse-graining and of linking coarse-grained models to the underlying small-scale physics that drive larger-scale dynamics. The core of this approach is an iterative refinement cycle, which moves between coarse-grained representations and all-atom representations. The coarse-grained models help to identify important regions of fine-grained phase space, while the fine-grained simulations allow the design and parameterization of better coarse-grained models. This research also involves the investigation and development of highly scalable, fault-tolerant modeling software to provide good simulation performance on modern heterogeneous supercomputers, for which individual processing units may be unreliable and the distribution of work across the processors can be severely unbalanced.
- Izvekov, S., & Voth, G. A. (2005). A multiscale coarse-graining method for biomolecular systems. The Journal of Physical Chemistry B, 109(7), 2469-73. doi:10.1021/jp044629q
- Noid, W. G., Chu, J.-W., Ayton, G. S., & Voth, G. A. (2007). Multiscale coarse-graining and structural correlations: connections to liquid-state theory. The Journal of Physical Chemistry B, 111(16), 4116-27. doi:10.1021/jp068549t
- Noid, W. G., Chu, J.-W., Ayton, G. S., Krishna, V., Izvekov, S., Voth, G. A., Das, A., & Andersen, H. C. (2008). The multiscale coarse-graining method. I. A rigorous bridge between atomistic and coarse-grained models. Journal Of Chemical Physics, 128(24), 244114
- Lyman, E., Pfaendtner, J., & Voth, G. A. (2008). Systematic multiscale parameterization of heterogeneous elastic network models of proteins. Biophysical Journal, 95(9), 4183-92. doi:10.1529/biophysj.108.139733
- Zhang, Z., Lu, L., Noid, W. G., Krishna, V., Pfaendtner, J., & Voth, G. A. (2008). A systematic methodology for defining coarse-grained sites in large biomolecules. Biophysical Journal, 95(11), 5073-83. doi:10.1529/biophysj.108.139626
Multiscale study of membrane-remodeling by proteins
Remodeling of biological membranes induced by proteins is an essential physiological process that facilitates key cellular tasks such as endocytosis, pathogen infection, immune response, cellular motility, protein trafficking, etc. By using advanced multiscale methods developed in our group, we combine all-atom, coarse-grained and continuum mechanics simulations in a powerful way to elucidate the molecular nature of membrane processes, as well as their large-scale physical and biophysical properties. Specifically, we focus on multiscale study of membrane remodeling induced by Bin/Amphiphysin/Rvs-homology (BAR) proteins and their connection to cellular mechanisms.
BAR proteins are essential modulators of the dynamics of cellular membranes. Given our recent advancements in developing multiscale techniques (FIGURE 1), we are in a unique position to understand the structural, biological, and physical properties of these processes at both the molecular and mesoscopic levels. Specifically, we study the molecular mechanism of BAR-mediated remodeling and how the molecular interactions couple with the long-wavelength behavior of biological membranes. Recently, our achievement in reconstructing molecular configurations from remodeled continuum membranes has allowed us to study the molecular details of very large remodeled vesicles (visible with optical microscopy!), which had not yet been studied theoretically.
Figure 1: An overview of the study of N-BAR mediated membrane remodeling. A) Remodeling of a liposome 400 nm in diameter, over the course of a single mesoscopic simulation. B) Connecting the two levels of resolution: the reticulated configurations at the continuum level are used to recreate a molecular representation of the system. C) Comparing the reticulated structures with electron microscopy (EM), with the aim to validate the model. The EM density maps will be used in reconstructing molecular representations.
Multiscale Simulation of G-protein-coupled receptors
G-protein-coupled receptors (GPCRs) are a superfamily of proteins with critical functions in cellular signal transduction, representing a primary class of drug targets. Since GPCRs are embedded in the membrane environment, obtaining their structural and dynamic information is generally difficult to achieve in experiments. Despite some success in X-ray crystallography, most GPCR structures remain unknown, while many questions about the ligand-dependent activation mechanism have not been answered. Thus elucidating these details from computational studies is very valuable. Using multiscale modeling techniques, we are developing GPCR models at hierarchical resolutions. Our fine-grained model bound to multiple agonists and antagonists provides mechanistic insights into the ligand-dependent GPCR activation beyond what has been found in crystal structures. Our mixed-resolution model, combining the fine-grained GPCR model and the coarse-grained environment, is being developed to enable highly accurate and efficient simulations and to provide a solution to GPCR homology models refinement and ligand screening. Our coarse-grained model, built on our fine-grained model, will help to reveal protein-protein interactions in GPCR oligomer and GPCR-G-protein complex. We believe that our study contribute to the understanding of the general GPCR molecular mechanism of action as well as the development of practical tools for GPCR structure-based drug discovery.
Figure 2: Adenosine A2A receptor is being studied as a prototypical GPCR. The fine-grained model (left) and the coarse-grained model (right).
- Mim, Carsten; Cui, Haosheng; Gawronski-Salerno, Joseph A; Frost, Adam; Lyman, Edward; Voth, Gregory A; Unger, Vinzenz M (2012). Structural basis of membrane bending by the N-BAR protein endophilin Cell, 149 (1) :137-45.
- Lyman, Edward; Cui, Haosheng; Voth, Gregory A (2010). Water under the BAR Biophysical Journal. 99 (6) :1783-90.
- Ayton, Gary S; Lyman, Edward; Krishna, Vinod; Swenson, Richard D; Mim, Carsten; Unger, Vinzenz M; Voth, Gregory A (2009). New insights into BAR domain-induced membrane remodeling Biophys. J., 97 (6) :1616-25.
- Izvekov, Sergei; Voth, Gregory A (2009). Solvent-free lipid bilayer model using multiscale coarse-graining J. Phys. Chem. B. 113 (13) :4443-55.
- Ayton, Gary S; Blood, Philip D; Voth, Gregory A (2007). Membrane remodeling from N-BAR domain interactions: insights from multi-scale simulation Biophys. J. 92 (10) :3595-602.
- Ayton, Gary S; Voth, Gregory A (2009). Hybrid coarse-graining approach for lipid bilayers at large length and time scales J. Phys. Chem. B. 113 (13) :4413-24.
- Ayton, Gary S; McWhirter, J Liam; McMurtry, Patrick; Voth, Gregory A (2005). Coupling field theory with continuum mechanics: a simulation of domain formation in giant unilamellar vesicles, Biophys. J. 88 (6) :3855-69.
Proteins and Protein Complexes
The Voth group is interested in applying computational methods to understand the self-assembly and dynamics of macromolecular assemblies, such as the HIV-1 viral capsid, networks of actin and actomyosin filaments, and the DNA helicase machinery, amongst others. The dynamics of these systems is complicated by their interactions with accessory proteins, for example in actin networks filament turnover is regulated by a variety of cofactors through mechanisms such as formin-mediated filament elongation and profilin-mediated nucleotide exchange.
These systems are difficult to simulate in part because traditional molecular dynamics provides insufficient sampling of high-energy states over the timescales accessible with current computational power. The large size, complex composition, and slow dynamics of these systems require the use of more advanced simulation techniques. Two of the main paradigms for overcoming this limitation are coarse-grained methods and accelerated sampling techniques.
In addition, these systems are interesting because the structure and dynamics of the small-scale components significantly impact the larger scale assembly and dynamics of the overall system. To address these types of multiscale problems, the Voth group, in collaboration with the rest of the Center for Multiscale Theory and Simulation, is developing novel ways of thinking about coarse-graining and of linking coarse-grained models to the underlying atomistic scale physics that drive protein dynamics. The core of this approach is an iterative refinement cycle, which moves between coarse-grained representations of the large-scale assemblies, and all-atom representations of the interactions between individual proteins (Figure, panel A). The coarse-grained models help to identify important interaction regions while the all-atom simulations allow the design and parameterization of better coarse-grained models. This research also involves the investigation and development of highly scalable modeling software to provide good simulation performance on modern supercomputers when the distribution of work across the CPUs can be severely unbalanced.
Figure 3: (A) Schematic of the iterative refinement cycle for multiscale coarse graining of biomolecular complexes. (B) An example coarse-grained model of the HIV-1 capsid protein in both the mature conical capsid form (B after Pornillos et al., Nature 469:424-428), and (C) as a mature-style protein lattice assembling on the surface of a sphere. (D) Profilin, a small protein which binds to actin monomers in the cytosol and accelerates nucleotide exchange, and formin, a large, multi-domain protein that works in conjunction with profilin to rapidly assemble actin filaments.
- Saunders, M.G. & Voth, G. A. (2012). Coarse-graining methods for computational biology. Ann. REv. Biophys. in press
- Grime, J. & Voth, G. A. (2012). Early stages of the HIV-1 capsid protein lattice formation. Biophys. J. in press.
- Saunders, M. G. & Voth, G. A. (2012). Comparison between actin filament models: coarse-graining reveals essential differences. Structure 20, 641.
- Ayton, G. S. & Voth, G. A. (2010). Multiscale computer simulation of the immature HIV-1 virion. Biophys. J. 99, 2757-65.
- Zhang, Z. & Voth, G. A. (2010). Coarse-Grained Representations of Large Biomolecular Complexes from Low-Resolution Structural Data. J. Chem. Theor. Comp. 6, 2990-3002.
- Saunders M. G. and G. A. Voth (2012). "Coarse-graining of multiprotein assemblies." Current Opinion in Structural Biology.
- Pfaendtner, J., N. Volkmann, et al. (2012). "Key structural features of the actin filament arp2/3 complex branch junction revealed by molecular simulation." Journal of molecular biology 416(1): 148-161.
- Zhang, Z. Y., K. Y. Sanbonmatsu, et al. (2011). "Key Intermolecular Interactions in the E. coli 70S Ribosome Revealed by Coarse-Grained Analysis." Journal of the American Chemical Society 133(42): 16828-16838.
- Pfaendtner, J., E. Lyman, et al. (2010). "Structure and dynamics of the actin filament." Journal of molecular biology 396(2): 252-263.
- Pfaendtner, J., E. M. De La Cruz, et al. (2010). "Actin filament remodeling by actin depolymerization factor/cofilin." Proceedings of the National Academy of Sciences of the United States of America 107(16): 7299-7304.
- Pfaendtner, J., D. Branduardi, et al. (2009). "Nucleotide-dependent conformational states of actin." Proceedings of the National Academy of Sciences of the United States of America 106(31): 12723-12728.
- Chu, J. W. and G. A. Voth (2006). "Coarse-grained modeling of the actin filament derived from atomistic-scale simulations." Biophys J 90(5): 1572-1582.
- Chu, J. W. and G. A. Voth (2005). "Allostery of actin filaments: molecular dynamics simulations and coarse-grained analysis." Proceedings of the National Academy of Sciences of the United States of America 102(37): 13111-13116.