Living systems achieve self-assembly of highly ordered and functional patterns with striking ease, control, and robustness. In contrast, experimental efforts to guide the assembly of nanoparticles into ordered materials with useful properties are often stymied by defective growth associated with severe kinetic traps, even when the desired structure is thermodynamically favorable.
The Geissler group dissects the energetics and dynamics that drive self-assembly of natural and engineered nanoscale systems. This work is often conducted in close collaboration with experimentalists, and we aim to provide practical guidelines for improving their control over or adjusting the outcome of the self-assembly process in the lab.
Some of our projects guide the creation of functional nanomaterials, including nanorod array formation for solar energy devices [Nano Lett 2010, Asaph]; simultaneous assembly of block copolymers and nanoparticles in interacting mixtures [Carl, Sucheol]; novel dense packings and stable open structures of polyhedra [Michael, Joseph]; and evaporation-driven assembly of magnetic nanoparticles [JACS 2011, JiYeon]. Other projects uncover the control parameters behind biological examples of self-assembly such as the competing aggregation modes of Archael chaperonin proteins [Nano Lett 2009, Carl, Steve], and the dynamic reorganization of light-harvesting proteins in photosynthetic membrane stacks [Anna]. To make our work computationally feasible, we develop coarse-grained models and advanced simulation techniques that accurately capture the salient thermodynamic and kinetic features of our systems [methods].
Complex biological motifs such as lipid membranes, actin networks, and DNA strands are not only functional components of living cellsthey can also be viewed as soft materials characterized by properties like bending rigidity, compressibility, or persistence length. The potentially large deformations of these biomaterials can produce surprising results, especially in the presence of an external driving force, yet the extent of the underlying fluctuations can be difficult to extract from even simple experimental probes.
The Geissler group provides a microscopic picture and a statistical-mechanical framework for interpreting biophysical experiments involving membranes, gels, and semiflexible filaments. We also develop the coarse-grained models, Monte Carlo moves, and other simulation techniques necessary to recapitulate biologically relevant energetics and dynamics [methods].
Our work on membranes and actin, a long-standing collaboration with the Fletcher group, has examined the growth of filopodium-like structures through coupled deformations of actin filaments and a membrane [Nature Physics 2008, PRL 2008, Lutz]; dynamics and elasticity of growing and strained actin gels [JCP 2009, Evan H, Sander]; curvature generation by membrane-binding proteins [Chris]; and the role of filament bending in actin branching [Evan W]. Other projects have extracted high-resolution information about DNA bending and stretching from single-molecule and ensemble experiments [Nano Lett 2007, JCP 2008, Biophys J 2008, Biophys J 2008, Biophys J 2009, PRE 2010, David, Steve].
Proteins fold, fluctuate, and respond to mechanical force on a complex free energy landscape, populating unfolded states and misfolded disease-associated traps on the way to the functional native state. Molecular evolutionary dynamics can also be posed as a nonequilibrium process on a rugged landscapegenotypic mutations over time give rise to variation in fitness measures via microscopic physical principles. Atomistic simulations can thoroughly characterize fast fluctuations about a particular native state, yet the long timescale and broad nature of the protein folding problem calls for generalizable coarse-grained models.
The Geissler group studies the thermodynamics and kinetics of protein fluctuation, folding, and evolution. One of our primary simulation frameworks is the Gō-like lattice heteropolymer model, a simple model that captures surprisingly complex features of the folding dynamics, free energy landscape, and force-dependent single-molecule experiments.
With the Gō-like model, we have probed the role non-native contacts play in the folding process, investigated the diversity of mechanisms by which a sequence folds to its native state, and examined the effect of evolutionary mutations on a sequence's ability to fold efficiently or resist mechanical unfolding [JMB 2009, Brian, Katie]. Other projects have used atomistic simulations to calculate the side-chain entropy in folded proteins, to relate correlated side-chain fluctuations to ligand binding and signal transduction, and to probe how branched side-chains inhibit secondary structure formation [JMB 2009, JCP 2010, Kateri, Will]. We also maintain collaborations with force spectroscopists and developers of Markov state models.
Recent simulations and experiments have confirmed that certain ions are present at air-water interfaces in enhanced concentrations, although dielectric continuum theory suggests that there should be strong forces driving ions away from those interfaces. Intriguingly, an ion's surface propensity is correlated with its position in the Hofmeister series, the highly specific ordering of ions according to the strength of their effects on various phenomena like surface tension and protein solubility. Elucidating that link will require examination of specific ion properties, and of general features of the behavior of polar solutes in inhomogeneous environments.
The Geissler group pursues a deep physical understanding of how polar solutes interact with interfaces through theory, simple models, and atomistic simulations. We also collaborate with practitioners of sum frequency generation and other interface-sensitive spectroscopies.
In simulations of liquid/vapor systems of both water and of the ideal polar Stockmayer fluid, we have shown that charged solutes prefer interfaces to bulk over a wide range of parameters; the surface propensity of ions in these two solvents exhibit the same sensitivity to ion size and charge; and electrostatic forces can actually stabilize ions at interfaces [PNAS 2009, Joyce, Pat]. Other projects have analyzed the behavior of ions or amino acids at interfaces via comparison to sum frequency generation [Israel J Chem 2007, JPCB 2009, CPL 2009, Joyce, Will].
Processes like self-assembly of nanoparticles, phase transformations of bulk and nanoscale materials, or conformational changes of biomolecules are characterized by largely disparate time and length scales. Thus, these phenomena can pose severe problems to straightforward computer simulation. The Geissler research group develops coarse-grained models and advanced simulation techniques that focus on the physically relevant parts of a system's evolution in time and space. In particular, the Geissler group has been involved in the development of the transition path sampling technique, successfully applying it to unravel the mechanisms of processes like autoionization in liquid water [Science 2001] and structural phase transformations in nanocrystals [JCP 2007]. We have also developed novel coarse-grained models for lipid membranes [JCP 2010] and patchy nanoparticles [JCP 2007]; and novel Monte Carlo moves for strained systems [JCP 2009] and self-assembling systems [Soft Matt 2009].