Free Energy Surfaces and Barriers for Vacancy Diffusion on Al(100), Al(110), Al(111) Reconstructed Surfaces

Metadynamics is a popular enhanced sampling method based on the recurrent application of a history-dependent adaptive bias potential that is a function of a selected number of appropriately chosen collective variables. In this work, using metadynamics simulations, we performed a computational study for the diffusion of vacancies on three different Al surfaces [reconstructed Al(100), Al(110), and Al(111) surfaces]. We explored the free energy landscape of diffusion and estimated the barriers associated with this process on each surface. It is found that the surfaces are unique regarding vacancy diffusion. More specically, the reconstructed Al(110) surface presents four metastable states on the free energy surface having sizable and connected passage-ways with an energy barrier of height 0.55 eV. On the other hand, the reconstructed Al(100)/Al(111) surfaces exhibit two/three metastable states, respectively, with an energy barrier of height 0.33 eV. The findings in this study can help to understand surface vacancy diffusion in technologically relevant Al surfaces.

Standard binding free energy and membrane desorption mechanism for a phospholipase C

Peripheral membrane proteins (PMPs) bind temporarily to cellular membranes and play important roles in signalling, lipid metabolism and membrane trafficking. Obtaining accurate membrane-PMP affinities using experimental techniques is more challenging than for protein-ligand affinities in aqueous solution. At the theoretical level, calculation of standard protein-membrane binding free energy using molecular dynamics simulations remains a daunting challenge owing to the size of the biological objects at play, the slow lipid diffusion and the large variation in configurational entropy that accompanies the binding process. To overcome these challenges, we used a computational framework relying on a series of potential-of-mean-force (PMF) calculations including a set of geometrical restraints on collective variables. This methodology allowed us to determine the standard binding free energy of a PMP to a phospholipid bilayer using an all-atom force field. Bacillus thuringiensis phosphatidylinositol-specific phospholipase C (BtPI-PLC) was chosen due to its importance as a virulence factor and owing to the host of experimental affinity data available. We computed a standard binding free energy of -8.2±1.4 kcal/mol in reasonable agreement with the reported experimental values (-6.6±0.2 kcal/mol). In light of the 2.3-μs separation PMF calculation, we investigated the mechanism whereby BtPI-PLC disengages from interactions with the lipid bilayer during separation. We describe how a short amphipathic helix engages in transitory interactions to ease the passage of its hydrophobes through the interfacial region upon desorption from the bilayer.

Discover, Sample and Refine: Exploring Chemistry with Enhanced Sampling Techniques

Over the last few decades enhanced sampling methods have made great strides. Here, we exploit this progress and propose a modular workflow for blind reaction discovery and characterization of reaction paths. Central to our strategy is the use of the recently developed explore variant of the on-the-fly probability enhanced sampling method. Like metadynamics, this method is based on the identification of appropriate collective variables. Our first step is the discovery of new chemical reactions and it is performed biasing a one dimensional collective variable derived from spectral graph theory. Once new reaction pathways are detected, we construct ad-hoc tailored neural-network based collective variables to improve sampling of specific reactions and finally we refine the results using free energy perturbation theory. Our workflow has been successfully applied to both intramolecular and intermolecular reactions. Without any chemical hypothesis, we discovered several possible products, computed the free energy surface at semiempirical level, and finally refined it with a more accurate Hamiltonian. Our workflow requires minimal user input, and thanks to its modularity and flexibility, can extend the scope of ab initio molecular dynamics for the exploration and characterization of reaction space.

Deeptime: a Python library for machine learning dynamical models from time series data

Generation and analysis of time-series data is relevant to many quantitative fields ranging from economics to fluid mechanics. In the physical sciences, structures such as metastable and coherent sets, slow relaxation processes, collective variables, dominant transition pathways or manifolds and channels of probability flow can be of great importance for understanding and characterizing the kinetic, thermodynamic and mechanistic properties of the system. Deeptime is a general purpose Python library offering various tools to estimate dynamical models based on time-series data including conventional linear learning methods, such as Markov state models (MSMs), Hidden Markov Models and Koopman models, as well as kernel and deep learning approaches such as VAMPnets and deep MSMs. The library is largely compatible with scikit-learn, having a range of Estimator classes for these different models, but in contrast to scikit-learn also provides deep Model classes, e.g. in the case of an MSM, which provide a multitude of analysis methods to compute interesting thermodynamic, kinetic and dynamical quantities, such as free energies, relaxation times and transition paths. The library is designed for ease of use but also easily maintainable and extensible code. In this paper we introduce the main features and structure of the deeptime software. Deeptime can be found under https://deeptime-ml.github.io/.

Optical solitons for the Kaup–Newell equation by collective variables method

In this work, we present a collective variable (CV) approach to establish dispersive solitary wave solutions for the Kaup–Newell Equation (KNE). The full CV theory has been utilized to enunciate the soliton molecules through its ground-laying parameters including the power of each pulse, phase and center-of-mass. Additionally, the dynamics of an ultra short pulse has been analyzed by using CV. This work may be utilized for various dynamics of solitons as well as the influence the amplitude, temporal position, frequency, phase and chirp on the solitons’ nonlinear parameters. Moreover, the numerical simulations have been designed by means of appropriate parameter values to explain more on the obtained results.

Binding Affinity Estimation From Restrained Umbrella Sampling Simulations

The protein-ligand binding affinity quantifies the binding strength between a protein and its ligand. Computer modeling and simulations can be used to estimate the binding affinity or binding free energy using data- or physics-driven methods or a combination thereof. Here, we discuss a purely physics-based sampling approach based on biased molecular dynamics (MD) simulations, which in spirit is similar to the stratification strategy suggested previously by Woo and Roux. The proposed methodology uses umbrella sampling (US) simulations with additional restraints based on collective variables such as the orientation of the ligand. The novel extension of this strategy presented here uses a simplified and more general scheme that can be easily tailored for any system of interest. We estimate the binding affinity of human fibroblast growth factor 1 (hFGF1) to heparin hexasaccharide based on the available crystal structure of the complex as the initial model and four different variations of the proposed method to compare against the experimentally determined binding affinity obtained from isothermal calorimetry (ITC) experiments. Our results indicate that enhanced sampling methods that sample along the ligand-protein distance without restraining other degrees of freedom do not perform as well as those with additional restraint. In particular, restraining the orientation of the ligands plays a crucial role in reaching a reasonable estimate for binding affinity. The general framework presented here provides a flexible scheme for designing practical binding free energy estimation methods.

Deep learning the slow modes for rare events sampling

The development of enhanced sampling methods has greatly extended the scope of atomistic simulations, allowing long-time phenomena to be studied with accessible computational resources. Many such methods rely on the identification of an appropriate set of collective variables. These are meant to describe the system’s modes that most slowly approach equilibrium under the action of the sampling algorithm. Once identified, the equilibration of these modes is accelerated by the enhanced sampling method of choice. An attractive way of determining the collective variables is to relate them to the eigenfunctions and eigenvalues of the transfer operator. Unfortunately, this requires knowing the long-term dynamics of the system beforehand, which is generally not available. However, we have recently shown that it is indeed possible to determine efficient collective variables starting from biased simulations. In this paper, we bring the power of machine learning and the efficiency of the recently developed on the fly probability-enhanced sampling method to bear on this approach. The result is a powerful and robust algorithm that, given an initial enhanced sampling simulation performed with trial collective variables or generalized ensembles, extracts transfer operator eigenfunctions using a neural network ansatz and then accelerates them to promote sampling of rare events. To illustrate the generality of this approach, we apply it to several systems, ranging from the conformational transition of a small molecule to the folding of a miniprotein and the study of materials crystallization.

Do Hospital Workers Feel They Are Ready to Manage a Sanitary Crisis in a Pre-Crisis Context?

Objective: The aim of this study was to measure the perception of readiness to manage a sanitary crisis for hospital workers and to study the factors related to this perception. Methods: This study was a cross-sectional study; 408 French hospital workers responded to an online questionnaire. The variables studied concerned the perceived personal preparedness, the perception of colleagues’ and hospital’s preparedness, perception of the situation, and preparatory behavioral acts. Correlations, partial correlations, and multiple linear regressions were applied. Results: Based on Pearson’s correlations, the higher the participants’ sense of personal efficacy and control over their behavior, the more ready they feel (r p = 0.77*** and r p = 0.55***). The more participants perceive their colleagues as ready and their hospital as prepared, the more ready they feel (r p = 0.52*** and r p = 0.46***). Based on Pearson’s partial correlations, upon controlling the effect of preparedness perception, declared preparedness is not significantly correlated with personal readiness perception (r p = 0.01). Conclusion: The perception of personal readiness does not depend only on actual preparedness but also on individual and collective variables. Technically, these results confirm the value of relying on psychosocial variables during training. It would be interesting to propose empowerment in training courses. It also seems necessary to demonstrate crisis management efficacy at different levels: institutional, collective, and individual.

Behavior of Linear and Nonlinear Dimensionality Reduction for Collective Variable Identi�cation of Small Molecule Solution-Phase Reactions

Identifying collective variables for chemical reactions is essential to reduce the 3$N$ dimensional energy landscape into lower dimensional basins and barriers of interest. However in condensed phase processes, the non-meaningful motions of bulk solvent often overpower the ability of dimensionality reduction methods to identify correlated motions that underpin collective variables. Yet solvent can play important indirect or direct roles in reactivity and much can be lost through treatments that remove or dampen solvent motion. This has been amply demonstrated within principal component analysis, although less is known about the behavior of nonlinear dimensionality reduction methods, e.g., UMAP, that have become more popular recently. The latter presents an interesting alternative to linear methods though often at the expense of interpretability. This work presents distance attenuated projection methods of atomic coordinates that facilitate the application of both PCA and UMAP to identify collective variables in solution, and further the specific identity of solvent molecules that participate in chemical reactions. The performance of both methods is examined in detail for two reactions where the explicit solvent plays very different roles within the collective variables. The first reaction consists of the dynamic exchange of a cation about a polyhydroxy anion that is facilitated by waters of solvation, while the second reaction consists of a nucleophilic attack of water upon ethylene to initiate cis/trans isomerization. When applied to raw data, both PCA and UMAP representations are dominated by bulk solvent motions. On the other hand, when applied to data preprocessed by our attenuated projection methods, both PCA and UMAP identify the appropriate collective variables in solution.