All living cells have a barrier that separates them from their environment: a thin self-assembled structure called a “lipid bilayer membrane”. Eukaryotic cells also have numerous membraneous structures inside themselves to further compartmentalize distinct organelles, such as the cell nucleus or the endoplasmic reticulum. It turns out that biomembranes play important roles in innumerable cellular processes, and that one key to their functional diversity lies in the remarkable elastic properties which they exhibit: lipid bilayers are molecularly thin two dimensional fluids which resists bending, stretching, and lipid tilting in such a way as to communicate forces and information through elasticity and geometry. In my talk I will discuss some of the fascinating aspects of lipid membrane biophysics, and I will explain in more detail several novel approaches we have developed in my group to probe both the various phenomena and the elastic properties of these amazing structures.
Digging Deep into Reactions with New First Principles Techniques
Effective sampling of reaction pathways is a longstanding challenge in molecular simulation. Difficulties in this area often result from chemical interactions between the environment and the reactive groups, resulting in a high dimensionality of the reaction pathway. In addition to this well-known issue of sampling environmental degrees of freedom, a second challenge is just as basic: realistic hypotheses of the reaction mechanism are needed, but not always available for systems of emerging interest. In this talk, I will introduce graph-based approaches that can thoroughly evaluate reaction coordinates in order to computationally locate the most kinetically feasible reaction pathways. Complementary advances in the Growing String Method for optimizing reaction paths at low effort will also be discussed. These tools are widely applicable to stoichiometric and catalytic systems that include simple environments and complex, solution-phase reactions. Examples will be provided for reactive systems where solvent was found to play an important role in the reaction. These will include catalysis involving transmetalation reactions for electronic polymer growth, and C-H activation reactions.
Excited States and Energy Conversion in Organic Crystals and at Interfaces via First-Principles Methods
Note the location change:
Harvard University, Mallinckrodt Building, Pfizer Lecture Hall
Organic crystals and hybrid interfaces are highly tunable, diverse
classes of cheap-to-process materials with promise for next-generation
optoelectronics. Further development of new materials requires new
intuition that links atomic- and molecular-scale morphology to
underlying excited-state properties and phenomena. I will review ab
initio methods for calculating excited-state and transport properties
of crystalline solids and interfaces, and show several applications,
where we have used these methods to explain or drive new experiments.
Specifically, I will cover the use of first-principles density
functional theory with tuned hybrid functionals, and many-body
perturbation theory within the GW approximation and the Bethe-Salpeter
equation approach, for computing and understanding spectroscopic
properties of acene crystals, including new insights into measured
multiexciton phenomena such as singlet fission; as time permits, I
will additionally share preliminary results on low-dimensional
materials, such as 2d chalcogenides, and halide perovskites. I will
also discuss multiple approaches to calculating level alignment at
metal-molecule interfaces, where we have recently generalized
optimally-tuned range-separated hybrid functionals to treat the
electronic structure with accuracy comparable to many-body
perturbation theory, and describe implications for single-molecule
junction transport measurements.
Insights into the structure and dynamics of biomolecules in cellular environments from computer simulations
Biological macromolecules such as proteins and nucleic acids have become well-understood at the single molecule level but it is much less clear how the structure-dynamics-function paradigms established largely under dilute and homogeneous conditions hold up under realistic biological conditions where crowding, heterogeneity, and the presence of a diverse set of metabolites are important factors. Using computational approaches we are exploring model systems of dense crowded systems ranging from simple spherical crowder models to concentrated protein solutions and a comprehensive model of a bacterial cytoplasm with all of the key components present in full atomistic detail. Simulations of these systems show altered dynamic properties, suggest the possibility of protein native state destabilization due to protein-protein and protein-metabolite interactions, altered solvent and metabolite behavior, and non-specific interactions between functionally related enzymes as a result of crowding. Some of the work described involves very large scale computer simulations that were enabled by methodological advances that will also be briefly discussed.
Exploration and learning of free energy landscapes of molecular crystals and oligopeptides
Theory, computation, and high-performance computers are playing an increasingly important role in helping us understand, design, and characterize a wide range of functional materials, chemical processes, and biomolecular/biomimetic structures. The synergy of computation and experiment is fueling a powerful approach to address some of the most challenging scientific problems. In this talk, I will describe the efforts we are making in my group to develop new computational methodologies that address specific challenges in free energy exploration and generation. In particular, I will describe our recent development of enhanced free energy based methodologies for predicting structure and polymorphism in molecular crystals and for determining conformational equilibria of oligopeptides. The strategies we are pursuing include heterogeneous multiscale modeling techniques, which allow “landmark” locations (minima and saddles) on a high-dimensional free energy surface to be mapped out, and temperature-accelerated methods, which allow relative free energies of the landmarks to be generated efficiently and reliably. I will then discuss new schemes for using machine learning techniques to represent and perform computations using multidimensional free energy surfaces and navigate chemical compound space in an effort to discover new compounds.