Karsten Kreis

Advantages and challenges in coupling an ideal gas to atomistic models in adaptive resolution simulations

In adaptive resolution simulations, molecular fluids are modeled employing different levels of resolution in different subregions of the system. When traveling from one region to the other, particles change their resolution on the fly. One of the main advantages of such approaches is the computational efficiency gained in the coarse-grained region. In this respect the best coarse-grained system to employ in the low resolution region would be the ideal gas, making intermolecular force calculations in the coarse-grained subdomain redundant. In this case, however, a smooth coupling is challenging due to the high energetic imbalance between typical liquids and a system of non-interacting particles. In the present work, we investigate this approach, using as a test case the most biologically relevant fluid, water. We demonstrate that a successful coupling of water to the ideal gas can be achieved with current adaptive resolution methods, and discuss the issues that remain to be addressed.

A unified framework for force-based and energy-based adaptive resolution simulations

Adaptive resolution schemes enable molecular dynamics simulations of liquids and soft matter employing two different resolution levels concurrently in the same setup. These methods are based on a position-dependent interpolation of either forces or potential energy functions. While force-based methods generally lead to non-conservative forces, energy-based ones include undesired force terms proportional to the gradient of the interpolation function. In this work we establish a so far missing bridge between these formalisms making use of the generalized Langevin equation, thereby providing a unifying framework to traditionally juxtaposed approaches to adaptive simulations.

Adaptive Resolution Simulations with Self-Adjusting High-Resolution Regions.

In adaptive resolution simulations, different regions of a simulation box are modeled with different levels of detail. Particles change their resolution on-the-fly when traveling from one subregion to the other. This method is particularly useful for studying multiscale systems in which effects on a broad range of length and time scales play a role. Until now, the geometry of the high-resolution region has been limited to simple geometries of spherical, cuboid, or cylindrical form, whose shape does not change during the simulation. However, many phenomena involve changes in size and shape of system components, for example, protein folding, polymer collapse, nucleation, and crystallization. In this work, we develop a scheme that uses a series of overlapping spheres to allow for an arbitrary division of space into domains of different levels of resolution. Furthermore, the geometry is automatically adjusted on-the-fly during the simulation according to changes in size and shape of, for example, a solvated macromolecule within the high-resolution region. The proposed approach is validated on liquid water. We then simulate the folding of an atomistically detailed polypeptide solvated in a shell of atomistic water that changes shape as the peptide conformation changes. We demonstrate that the peptide folding process is unperturbed by the use of our methodology.

From Classical to Quantum and Back: A Hamiltonian Scheme for Adaptive Multiresolution Classical/Path-Integral Simulations

Quantum delocalization of atomic nuclei affects the physical properties of many hydrogen-rich liquids and biological systems even at room temperature. In computer simulations, quantum nuclei can be modeled via the path-integral formulation of quantum statistical mechanics, which implies a substantial increase in computational overhead. By restricting the quantum description to a small spatial region, this cost can be significantly reduced. Herein, we derive a bottom-up, rigorous, Hamiltonian-based scheme that allows molecules to change from quantum to classical and vice versa on the fly as they diffuse through the system, both reducing overhead and making quantum grand-canonical simulations possible. The method is validated via simulations of low-temperature parahydrogen. Our adaptive resolution approach paves the way to efficient quantum simulations of biomolecules, membranes, and interfaces.

Classifying, quantifying, and witnessing qudit-qumode hybrid entanglement

Recently, several hybrid approaches to quantum information emerged which utilize both continuous- and discrete-variable methods and resources at the same time. In this work, we investigate the bipartite hybrid entanglement between a finite-dimensional, discrete-variable quantum system and an infinite-dimensional, continuous-variable quantum system. A classification scheme is presented leading to a distinction between pure hybrid entangled states, mixed hybrid entangled states (those effectively supported by an overall finite-dimensional Hilbert space), and so-called truly hybrid entangled states (those which cannot be described in an overall finite-dimensional Hilbert space). Examples for states of each regime are given and entanglement witnessing as well as quantification are discussed. In particular, using the channel map of a thermal photon noise channel, we find that true hybrid entanglement naturally occurs in physically important settings. Finally, extensions from bipartite to multipartite hybrid entanglement are considered.

The relative entropy is fundamental to adaptive resolution simulations

Adaptive resolution techniques are powerful methods for the efficient simulation of soft matter systems in which they simultaneously employ atomistic and coarse-grained (CG) force fields. In such simulations, two regions with different resolutions are coupled with each other via a hybrid transition region, and particles change their description on the fly when crossing this boundary. Here we show that the relative entropy, which provides a fundamental basis for many approaches in systematic coarse-graining, is also an effective instrument for the understanding of adaptive resolution simulation methodologies. We demonstrate that the use of coarse-grained potentials which minimize the relative entropy with respect to the atomistic system can help achieve a smoother transition between the different regions within the adaptive setup. Furthermore, we derive a quantitative relation between the width of the hybrid region and the seamlessness of the coupling. Our results do not only shed light on the what and how of adaptive resolution techniques but will also help setting up such simulations in an optimal manner.

From classical to quantum and back: Hamiltonian adaptive resolution path integral, ring polymer, and centroid molecular dynamics

Path integral-based methodologies play a crucial role for the investigation of nuclear quantum effects by means of computer simulations. However, these techniques are significantly more demanding than corresponding classical simulations. To reduce this numerical effort, we recently proposed a method, based on a rigorous Hamiltonian formulation, which restricts the quantum modeling to a small but relevant spatial region within a larger reservoir where particles are treated classically. In this work, we extend this idea and show how it can be implemented along with state-of-the-art path integral simulation techniques, including path-integral molecular dynamics, which allows for the calculation of quantum statistical properties, and ring-polymer and centroid molecular dynamics, which allow the calculation of approximate quantum dynamical properties. To this end, we derive a new integration algorithm that also makes use of multiple time-stepping. The scheme is validated via adaptive classical-path-integral simulations of liquid water. Potential applications of the proposed multiresolution method are diverse and include efficient quantum simulations of interfaces as well as complex biomolecular systems such as membranes and proteins.