Search

This paper considers the narrow escape problem of a Brownian particle within a three‐dimensional Riemannian manifold under the influence of the force field. We compute an asymptotic expansion of mean sojourn time for Brownian particles. As a auxiliary result, we obtain the singular structure for the restricted Neumann Green's function which may be of independent interest. [ABSTRACT FROM AUTHOR]

Pupil diameter, under constant illumination, is known to reflect individuals' internal states, such as surprise about observation and environmental uncertainty. Despite the growing use of pupillometry in cognitive learning studies as an additional measure for examining internal states, few studies have used pupillometry in human motor learning studies. Here, we provide the first detailed characterization of pupil diameter changes in a short-term reach adaptation paradigm. We measured pupil changes in 121 human participants while they adapted to abrupt, gradual, or switching force field conditions. Sudden increases in movement error caused by the introduction/reversal of the force field resulted in strong phasic pupil dilation during movement accompanied by a transient increase in tonic premovement baseline pupil diameter in subsequent trials. In contrast, pupil responses were reduced when the force field was gradually introduced, indicating that large, unexpected errors drove the changes in pupil responses. Interestingly, however, error-induced pupil responses gradually became insensitive after experiencing multiple force field reversals. We also found an association between baseline pupil diameter and incidental knowledge of the gradually introduced perturbation. Finally, in all experiments, we found a strong co-occurrence of larger baseline pupil diameter with slower reaction and movement times after each rest break. Collectively, these results suggest that tonic baseline pupil diameter reflects one's belief about environmental uncertainty, whereas phasic pupil dilation during movement reflects surprise about a sensory outcome (i.e., movement error), and both effects are modulated by novelty. Our results provide a new approach for nonverbally assessing participants' internal states during motor learning. [ABSTRACT FROM AUTHOR]

The longevity of asphalt pavements is a key focus of road engineering, which closely relates to the self-healing ability of bitumen. Our work aims to establish a CGMD model and matched force field for bitumen and break through the limitations of the research scale to further explore the microscopic mechanism of bitumen self-healing. In this study, a CGMD mapping scheme containing 16 kinds of beads is proposed, and the non-bond potential energy function and bond potential energy function are calculated based on all-atom simulation to construct and validate a coarse-grained model for bitumen. On this basis, a micro-crack model with a width of 36.6nm is simulated, and the variation laws of potential energy, density, diffusion coefficient, relative concentration and temperature in the process of bitumen self-healing are analyzed with the cracking rate parameter proposed to characterize the degree of bitumen crack healing. The results show that the computational size of the coarse-grained simulation is much larger than that of the all-atom, which can explain the self-healing mechanism at the molecular level. In the self-healing process, non-bonded interactions dominate the molecular movement, and differences in the decreased rate of diffusion among the components indicate that saturates and aromatics play a major role in self-healing. Meanwhile, the variations in crack rates reveal that healing time is inversely proportional to temperature. The impact of increasing temperature on reducing healing time is most obvious when the temperature approaches the glass transition temperature (300 K). [ABSTRACT FROM AUTHOR]

Motivated by the problem of optimization of force-field systems in physics using large-scale computer simulations, we consider exploration of a deterministic complex multivariate response surface. The objective is to find input combinations that generate output close to some desired or “target” vector. Despite reducing the problem to exploration of the input space with respect to a 1-D loss function, the search is nontrivial and challenging due to infeasible input combinations, high dimensionalities of the input and output space and multiple “desirable” regions in the input space, and the difficulty of emulating the objective function well with a surrogate model. We propose an approach that is based on combining machine learning techniques with smart experimental design ideas to locate multiple good regions in the input space. Note to Practitioners—ReaxFF is a force field that incorporates complex functions with associated inputs in order to describe the inter- and intra-atomic interactions in materials systems. A typical ReaxFF force field consists of hundreds of parameters (inputs) per element type. During the development of a force field for a molecular system of interest, using computer simulations, these parameters are optimized to reproduce hundreds of material properties close to some benchmark reference values. Finding “good” combinations of hundreds of parameters that produce hundreds of reference values close to their gold standards is a challenging problem because there may be several parameter combinations that may be “almost equally good” or “equally desirable.” To add to the complication, several input combinations simply lead to a system crash, not producing any output at all. Standard global optimization methods do not address such a problem. We propose a novel framework that can address this problem. Beyond the ReaxFF optimization, it can be applied to multiobjective optimization in engineering and the physical sciences, where there are unknown constraints and the focus is on obtaining several good points that can serve as alternatives to a single global optimum. [ABSTRACT FROM AUTHOR]

Despite the massive application of end-point free energy methods in protein–ligand and protein–protein interactions, computational understandings about their performance in relatively simple and prototypical host–guest systems are limited. In this work, we present a comprehensive benchmark calculation with standard end-point free energy techniques in a recent host–guest dataset containing 13 host–guest pairs involving the carboxylated-pillar[6]arene host. We first assess the charge schemes for solutes by comparing the charge-produced electrostatics with many ab initio references, in order to obtain a preliminary albeit detailed view of the charge quality. Then, we focus on four modelling details of end-point free energy calculations, including the docking procedure for the generation of initial condition, the charge scheme for host and guest molecules, the water model used in explicit-solvent sampling, and the end-point methods for free energy estimation. The binding thermodynamics obtained with different modelling schemes are compared with experimental references, and some practical guidelines on maximizing the performance of end-point methods in practical host–guest systems are summarized. Further, we compare our simulation outcome with predictions in the grand challenge and discuss further developments to improve the prediction quality of end-point free energy methods. Overall, unlike the widely acknowledged applicability in protein–ligand binding, the standard end-point calculations cannot produce useful outcomes in host–guest binding and thus are not recommended unless alterations are performed. [ABSTRACT FROM AUTHOR]

The 3D-reference interaction site model (3D-RISM) molecular solvation theory in combination with the Kovalenko–Hirata (KH) closure is extended to seven heterocyclic liquids to understand their liquid states and to test the performance of the theory in solvation free energy (SFE) calculations of solutes in select solvents. The computed solvent site distribution profiles were compared with the all-atom molecular dynamics (MD) simulations, showing comparable performances. The computational results were compared against the structural parameters for liquids, whenever available, as well as against the experimental SFEs. The liquids are found to have local ordered structures held together via weak interactions in both the RISM and MD simulations. The 3D-RISM-KH computed SFEs are in good agreement with the benchmark values for the tetrahydrothiophene-S,S-dioxide, and showed comparatively larger deviations in the case of the SFEs in the tetrahydrofuran continuum. [ABSTRACT FROM AUTHOR]

Phosphorylation plays a key role in plant biology, such as the accumulation of plant cells to form the observed proteome. Statistical analysis found that many phosphorylation sites are located in disordered regions. However, current force fields are mainly trained for structural proteins, which might not have the capacity to perfectly capture the dynamic conformation of the phosphorylated proteins. Therefore, we evaluated the performance of ff03CMAP, a balanced force field between structural and disordered proteins, for the sampling of the phosphorylated proteins. The test results of 11 different phosphorylated systems, including dipeptides, disordered proteins, folded proteins, and their complex, indicate that the ff03CMAP force field can better sample the conformations of phosphorylation sites for disordered proteins and disordered regions than ff03. For the solvent model, the results strongly suggest that the ff03CMAP force field with the TIP4PD water model is the best combination for the conformer sampling. Additional tests of CHARMM36m and FB18 force fields on two phosphorylated systems suggest that the overall performance of ff03CMAP is similar to that of FB18 and better than that of CHARMM36m. These results can help other researchers to choose suitable force field and solvent models to investigate the dynamic properties of phosphorylation proteins. [ABSTRACT FROM AUTHOR]

Individuals with physical limb disabilities are often restricted to perform activities of daily life (ADLs). While efficacy of bilateral training has been demonstrated in improving physical coordination of human limbs, few robots have been developed in simulating people’s ADLs integrated with task-specific force field control. This study sought to develop a bilateral robot for better task rendering of general ADLs (gADLs), where gADL-consistent workspace is achieved by setting linear motors in series, and haptic rendering of multiple bimanual tasks (coupled, uncoupled and semi-coupled) is enabled by regulating force fields between robotic handles. Experiments were conducted with human users, and our results present a viable method of a single robotic system in simulating multiple physically bimanual tasks. In future, the proposed robotic system is expected to be serving as a coordination training device, and its clinical efficacy will be also investigated.

Molten salt nuclear reactors are promising technologies for sustainable energy sources. In this context, understanding the physicochemical properties of molten salts is essential to the development, usage, and maintenance of reactors. Classical molecular simulations allow estimating the macroscopic properties of molten salts at a relatively low computational cost. In this work, we develop a new force field for plutonium-based molten salts. We validate our classical model by comparing molecular dynamics results with experimental measurements, obtaining a quantitative agreement. Our work demonstrates the reliability and transferability of classical force fields to represent the physicochemical properties of molten salts. [ABSTRACT FROM AUTHOR]

The main cysteine protease (Mpro) of coronavirus SARS-CoV-2 has become a promising target for computational development in anti-COVID-19 treatments. Here, we benchmarked the performance of six biomolecular molecular dynamics (MD) force fields (OPLS-AA, CHARMM27, CHARMM36, AMBER03, AMBER14SB and GROMOS G54A7) and three water models (TIP3P, TIP4P and SPC) for reproducing the native fold and the enzymatic activity of Mpro as monomeric and dimeric units. The MD sampling up to 1 μs suggested that the proper choice of the force fields and water models plays an essential role in reproducing the tertiary structure and the inter-residue distance between the catalytic dyad His41-Cys145. We found that while most benchmarked all-atom force fields reproduce well the native fold of Mpro, the CHARMM27/TIP3P and OPLS-AA/TIP4P setups revealed a good performance in reproducing the structure of the catalytic domain. In addition, these FF setups were also well-adopted for MD sampling of Mpro at the physiologic conditions by mimicking the presence of 100 mM NaCl and the elevated temperature of 310 K. Finally, both FFs were also performed well in reproducing the native fold of Mpro in a dimeric form. Therefore, comparing the preservation of the native fold of Mpro and the stability of its catalytic site architecture, our MD benchmarking suggests that the OPLS-AA/TIP4P and CHARMM27/TIP3P MD setups at the physiologic conditions may be well-suited for rapid in silico screening and developing broad-spectrum anti-coronaviral therapeutic agents. [Display omitted] • The main cysteine protease (Mpro) of coronavirus SARS-CoV-2 is a promising target. • Benchmarking six MD force fields, such as (OPLS-AA, CHARMM27, CHARMM36, AMBER03, AMBER14SB and GROMOS G54A7, is performed. • Native fold of monomeric and dimeric Mpro is considered. • Tertiary structure and the inter-residue distance between the catalytic dyad His41-Cys145 as key parameters. [ABSTRACT FROM AUTHOR]

[Display omitted] • Yttria: metal oxide useful in technological applications. • 3-Aminopropyltriethoxy silane (APTES) is a suitable molecular system used as substrate for binding monoclonal antibodies to the surface of yttria nanoparticles. • Use of MD and DFT calculations to explain the interactions between yttria and APTES. • Chemisorption and physisorption of APTES on yttria surface. • Dative bond between nitrogen and yttrium atoms. In this paper, the development of Dreiding force field parameters to describe the structural features of yttria, has been done through a test and trial procedure. The parameters are used to study the interactions occurring between yttria's surface with 3-aminopropyltriethoxy silane (APTES) molecular system, which has been used as a substrate to bind a monoclonal antibody (anti-CYFRA-21-1) to an adequately prepared surface of yttria nanoparticles. The work reveals that, on an eventual hydroxyl free metal-oxide surfaces the APTES molecules would adsorb on it with two layers. The first is a chemisorbed layer that covers the entire surface (surface-density of 2.10 × 10−6 mol/m 2) with a direct partially covalent dative bond between the nitrogen of the amine group of APTES and the yttrium atoms on the surface. The second is a physisorbed layer over the first, with a half the surface-density (c.a. 1.05 × 10−6 mol/m 2), where the APTES molecules amine groups are pointing outwards with respect to the surface. This makes possible the direct covalent bonding between the amine groups, of the outer adsorbed layer of APTES molecules, with the carboxylic group of the monoclonal anti-CYFRA-21-1 antibody. Quantum mechanical calculations confirmed the outcome of the force field study. [ABSTRACT FROM AUTHOR]

It has been increasingly common to investigate dynamic and thermodynamic properties of green solvents at atomistic scales with molecular simulation. In this work, we present a detailed evaluation of the widely used fixed-charge regime, i.e., the combination of scaled RESP charges and GAFF derivatives. The benchmark set contains three ionic liquids formed by cations 1-butyl-3-methylimidazolium and 1-hexyl-3-methylimidazolium and anions bis(trifluoromethylsulfonyl)imide and PF6. For the charge scaling issue, two physical properties including the mass density and the solvation free energies of external agents in ionic liquids are selected as the criteria. Large-scale fast-growth solvation free energy simulations are performed to obtain the solvation thermodynamics, and the results are further combined with hydration data to form a water-ionic-liquids transfer dataset, which considers the partition of solutes between water and ionic liquids. It is observed that the density-derived charge scaling factor is always smaller than that derived from solvation/partition thermodynamics, which supports the use of a slightly larger scaling factor to obtain a balanced description of solute-solvent and solvent-solvent interactions. For the bonded-terms assessment, we refit the transferable GAFF2 derivatives with generalized force-matching in a molecule- and component-specific manner. The results suggest that the bond-stretching and angle-bending terms in GAFF derivatives are often problematic, while the torsional potential shows satisfactory reproduction of ab initio results. Finally, combining the extensive computational perspectives accumulated in our series works, general guidelines for molecular modelling of ionic liquids with fixed-charge force fields are summarized.

In molecular modelling of novel solvents such as ionic liquids, it is common to scale atomic charges to improve the experiment‐simulation agreement for some selected properties. As these liquids are designed to solvate solutes, whether the solvation thermodynamics could be correctly described is of utmost importance. Therefore, we present a comprehensive large‐scale calculation of solvation free energies via nonequilibrium fast‐switching simulations for a spectrum of molecules in ionic liquids, the atomic charges of which are scaled to maximize the prediction‐experiment correlation. Further, the density‐derived choice is compared with the solvation‐thermodynamics‐derived one. When the scaling factor is decreased, the density exhibits a monotonically decreasing behavior due to weaker inter‐molecular interactions produced by scaled atomic charges. However, solvation free energies do not show consistent monotonic behaviors, which are caused by competing electrostatic and vdW responses to the scaling‐parameter variation. More intriguingly, the solvation‐free‐energy‐derived scaling factor is generally slightly higher than the density‐derived one. We further calculate partition coefficients ortransfer free energies of solutes from water to ionic liquids to provide another thermodynamic perspective of charge scaling. Another central result is the detailed evaluation of the widely used force fields for bonded and vdW terms, i.e., the GAFF derivatives. [ABSTRACT FROM AUTHOR]

Molecular docking is one of the most popular computational tools for the hit discovery step in drug design. However, there is ample room for improvement of docking's ability to identify correct binding modes and discriminate active from decoy compounds. Molecular dynamics (MD) simulations of protein–ligand docking structures have been shown to be effective in improving docking results. Here, we present CHARMM‐GUI high‐throughput simulator (HTS) that prepares MD simulation systems and inputs for multiple protein–ligand complex structures in a high‐throughput manner. HTS supports commonly used MD programs (NAMD, GROMACS, AMBER, OpenMM, GENESIS, Desmond, LAMMPS, and Tinker) along with various force field combinations for protein and ligand, including CHARMM36m, Amber (ff19SB/ff14SB), OPLS‐AA/M, CGenFF, GAFF2, and OpenFF. Validation tests using Miller and the directory of useful decoys‐enhanced (DUD‐E) datasets demonstrate that short MD simulations using HTS‐generated systems and simple ligand RMSD calculations consistently outperform docking results. Specifically, MD simulations can better identify correct ligand‐binding modes among top 10 binding poses as compared to docking scores. In addition, MD simulations can better discriminate active from decoy compounds in the DUD‐E dataset than docking scores for both soluble and membrane proteins. We expect that HTS can be a useful tool to facilitate the hit discovery process in drug design by improving docking results. [ABSTRACT FROM AUTHOR]

Infrared (IR) spectroscopy is a valuable tool to obtain information about protein secondary structure. The far-infrared (FIR) spectrum is characterized by a complex combination of different molecular contributions which, for small molecules, may be interpreted with the help of quantum-mechanical (QM) calculations. Unfortunately, the high computational cost of QM calculations makes them inapplicable to larger molecules, such as proteins and peptides. In this work, we present a theoretical study on the secondary structure, molecular properties, and vibrational spectra of different peptides, using both a classical and a QM approach. Our results show that the amide I main peak value, and related quantities, such as dipole strength (DS) and transition dipole moment (TDM), depends on protein secondary structure; in particular, from QM calculations arises that α -rich molecular systems present lower intensities than β -rich ones. Furthermore, it is possible to decouple and identify the intensity of the different contributions of the inter- and intra-molecular motions which characterize the FIR spectrum, starting from the results obtained with QM calculations. [ABSTRACT FROM AUTHOR]

In this work, Layered Double Hydroxide (LDH) materials carrying the worldwide administered non-steroidal anti-inflammatory drug naproxen (NAP), and the sodium naproxenate salt (NaNAP) for comparison, were studied by computational approaches aiming to model the structure of hybrid LDH-drug and shed light on NAP intercalation process. Atomic modeling calculations were performed at the quantum mechanical level based on Density Functional Theory and classical force fields based on empirical interatomic potentials. LDH NAP materials were prepared by ion exchange reaction from Mg 2 Al(OH) 6 Cl and Zn 2 Al(OH) 6 Cl pristine phases. The characterization of the materials confirmed NAP intercalation and also the permanence of the pristine phases in the isolated materials after ion exchange. Crystallographic lattice parameters, elemental analysis, and TGA experimental results were then employed in the calculations, which revealed that NAP anions can completely neutralize the positive charge of the LDH layers: both Mg 2 Al and Zn 2 Al LDH structures could be optimized with all Cl− anions substituted by NAP. The drug assumed different dispositions in the NaNAP crystal or when intercalated into LDH. Additionally, infrared wavenumbers calculations agreed with the experimental results and showed useful to support LDH NAP bands assignment. The employed theoretical models to represent the structure of LDH NAP systems are expected to assist the interpretation of future experimental results and to be used as auxiliary tools to tune properties of LDH-drug pharmaceutical formulations. [Display omitted] • Density functional theory and classical force fields based on empirical interatomic potentials were applied for the Mg 2 Al-NAP and Zn 2 Al-NAP LDH materials. • The crystal structure of the NaNAP was optimized and the main IR wavenumbers were calculated and agreed with the experimental results. • Calculations showed that the charge of the LDH layers can be completely neutralized by NAP, thus achieving maximum ion exchange capacity. • NAP assumed different dispositions in the NaNAP crystal or intercalated into LDH. [ABSTRACT FROM AUTHOR]

Metal–organic frameworks (MOFs) are a highly versatile group of porous materials suitable for a broad range of applications, which often crucially depend on the MOFs' heat transport properties. Nevertheless, detailed relationships between the chemical structure of MOFs and their thermal conductivities are still largely missing. To lay the foundations for developing such relationships, we performed non-equilibrium molecular dynamics simulations to analyze heat transport in a selected set of materials. In particular, we focus on the impact of organic linkers, the inorganic nodes and the interfaces between them. To obtain reliable data, great care was taken to generate and thoroughly benchmark system-specific force fields building on ab-initio-based reference data. To systematically separate the different factors arising from the complex structures of MOF, we also studied a series of suitably designed model systems. Notably, besides the expected trend that longer linkers lead to a reduction in thermal conductivity due to an increase in porosity, they also cause an increase in the interface resistance between the different building blocks of the MOFs. This is relevant insofar as the interface resistance dominates the total thermal resistance of the MOF. Employing suitably designed model systems, it can be shown that this dominance of the interface resistance is not the consequence of the specific, potentially weak, chemical interactions between nodes and linkers. Rather, it is inherent to the framework structures of the MOFs. These findings improve our understanding of heat transport in MOFs and will help in tailoring the thermal conductivities of MOFs for specific applications. [ABSTRACT FROM AUTHOR]

• Repurposing existing drugs for new diseases is cost effective and time saving. • In silico methods are crucial for rapid drug screening in the early stages. • Machine learning algorithms confer speed and accuracy to computational screening approaches. • Deep learning is immensely powerful to design molecules with desired properties. The current global health emergency in the form of the Coronavirus 2019 (COVID-19) pandemic has highlighted the need for fast, accurate, and efficient drug discovery pipelines. Traditional drug discovery projects relying on in vitro high-throughput screening (HTS) involve large investments and sophisticated experimental set-ups, affordable only to big biopharmaceutical companies. In this scenario, application of efficient state-of-the-art computational methods and modern artificial intelligence (AI)-based algorithms for rapid screening of repurposable chemical space [approved drugs and natural products (NPs) with proven pharmacokinetic profiles] to identify the initial leads is a powerful option to save resources and time. Structure-based drug repurposing is a popular in silico repurposing approach. In this review, we discuss traditional and modern AI-based computational methods and tools applied at various stages for structure-based drug discovery (SBDD) pipelines. Additionally, we highlight the role of generative models in generating molecules with scaffolds from repurposable chemical space. [ABSTRACT FROM AUTHOR]

Deep eutectic solvents (DESs) are binary or ternary mixtures of compounds that possess significant melting point depressions relative to the pure isolated components. The discovery of DESs has been a major breakthrough with multiple fields benefitting from their low cost and tunable physiochemical properties. However, tailoring DESs for specific applications through their practically unlimited synthetic combinations can be as much a hindrance as a benefit given the expense and time‐required to perform large‐scale experimental measurements. This emphasizes the need for fast computational tools capable of making accurate predictions of DES physiochemical properties exclusively from molecular structure. Yet, these systems are not trivial to model or simulate at the atomic level given their exceedingly nonideal behaviors, asymmetry of components, and the complexity of their molecular electrostatic interactions. Despite the challenge, computational reports featuring quantum mechanical (QM) methods have provided significant understanding into the relationship between the melting point depression and the unique and complex hydrogen bond network present in DESs. Classical molecular dynamics (MD) methods have examined bulk‐phase solvent organization in conjunction with thermodynamic and transport properties. Machine learning (ML) algorithms have shown great potential as structure–property prediction tools. Overall, this review highlights computational accomplishments that have meaningfully advanced our understanding of DESs and strives to give the reader a sense of the overall strengths and drawbacks of the methodologies employed while hinting at promises of advances to come. This article is categorized under:Software > Simulation Methods [ABSTRACT FROM AUTHOR]

Metal-organic frameworks (MOFs) represent a distinctive class of nanoporous materials with considerable potential across a wide range of applications. Recently, a handful of MOFs has been explored for the storage of environmentally hazardous fluorinated gases (Keasler et al. Science 2023, 381, 1455), yet the potential of over 100,000 MOFs for this specific application has not been thoroughly investigated, particularly due to the absence of an established force field. In this study, we develop an accurate force field for nonaversive hydrofluorocarbon vinylidene fluoride (VDF) and conduct high-throughput computational screening to identify top-performing MOFs with high VDF adsorption capacities. Quantitative structure-property relationships are analyzed via machine learning models on the combinations of geometric, chemical, and topological features, followed by feature importance analysis to probe the effects of these features on VDF adsorption. Finally, from detailed structural analysis via radial distribution functions and spatial densities, we elucidate the significance of different interaction modes between VDF and metal nodes in top-performing MOFs. By synergizing force-field development, computational screening, and machine learning, our findings provide microscopic insights into VDF adsorption in MOFs that will advance the development of new nanoporous materials for high-performance VDF storage or capture.

More than a half century ago it became feasible to simulate, using classical-mechanical equations of motion, the dynamics of molecular systems on a computer. Since then classical-physical molecular simulation has become an integral part of chemical research. It is widely applied in a variety of branches of chemistry and has significantly contributed to the development of chemical knowledge. It offers understanding and interpretation of experimental results, semiquantitative predictions for measurable and nonmeasurable properties of substances, and allows the calculation of properties of molecular systems under conditions that are experimentally inaccessible. Yet, molecular simulation is built on a number of assumptions, approximations, and simplifications which limit its range of applicability and its accuracy. These concern the potential-energy function used, adequate sampling of the vast statistical-mechanical configurational space of a molecular system and the methods used to compute particular properties of chemical systems from statistical-mechanical ensembles. During the past half century various methodological ideas to improve the efficiency and accuracy of classical-physical molecular simulation have been proposed, investigated, evaluated, implemented in general simulation software or were abandoned. The latter because of fundamental flaws or, while being physically sound, computational inefficiency. Some of these methodological ideas are briefly reviewed and the most effective methods are highlighted. Limitations of classical-physical simulation are discussed and perspectives are sketched.

In this work we have derived the parameters of an AMOEBA-like polarizable forcefield for electrolytes based on tetramethoxy and tetraethoxy-glyoxal acetals, and propylene carbonate. The resulting forcefield has been validated using both ab-initio data and the experimental properties of the fluids. Using molecular dynamics simulations, we have investigated the structural features and the solvation properties of both the neat liquids and of the corresponding 1 M LiTFSI electrolytes at the molecular level. We present a detailed analysis of the Li ion solvation shells, of their structure and highlight the different behavior of the solvents in terms of their molecular structure and coordinating features., (© 2024 The Authors. ChemistryOpen published by Wiley-VCH GmbH.)

Despite the successes of antibodies as therapeutic binding proteins, they still face production and design challenges. Alternative binding scaffolds of smaller size have been developed to overcome these issues. A subset of these alternative scaffolds recognizes target molecules through mutations to a set of surface resides, which does not alter their backbone structures. While the computational design of antibodies for target epitopes has been explored in depth, the same has not been done for alternative scaffolds. The commonly used dock-and-mutate approach for binding proteins, including antibodies, is limited because it uses a constant sequence and structure representation of the scaffold. Docking fixed-backbone scaffolds with a varied group of surface amino acids increases the chances of identifying superior starting poses that can be improved with subsequent mutations. In this work, we have developed MutDock, a novel computational approach that simultaneously docks and mutates fixed backbone scaffolds for binding a target epitope by identifying a minimum number of hydrogen bonds. The approach is broadly divided into two steps. The first step uses pairwise distance alignment of hydrogen bond-forming areas of scaffold residues and compatible epitope atoms. This step considers both native and mutated rotamers of scaffold residues. The second step mutates clashing variable interface residues and thermodynamically unfavorable residues to create additional strong interactions. MutDock was used to dock two scaffolds, namely, Affibodies and DARPins, with ten randomly selected antigens. The energies of the docked poses were minimized and binding energies were compared with docked poses from ZDOCK and HADDOCK. The top MutDock poses consisted of higher and comparable binding energies than the top ZDOCK and HADDOCK poses, respectively. This work contributes to the discovery of novel binders based on smaller-sized, fixed-backbone protein scaffolds.

This research focuses on obtaining the behavior of chemical compounds with respect to their molecular weight and optimization energy based on the variation in properties in organic carbon links. Here, behavioral analysis of compounds is used in the application of a metal organic framework to denote the high-grade compounds. The grade was selected based on the essential measure of optimization energy and molecular weight, and in turn, depicts the stability of material. Computation of the optimization energy and molecular weight of chemical compounds was performed with Avogadro software. Several force fields can be considered to compute optimized energy. Exclusively, three force fields, namely, the Universal Force Field (UFF), the General Amber Force Field (GAFF), and the Ghemical force field (Ghemical) were selected from Avogadro as these were more relevant to compounds considered in this research. The various chemical compounds examined in this work are Aluminum (Al), Boron (Br), Calcium (Ca), Chlorine (Cl), Indium (In), Potassium (K), Scandium (Sc), Silicon (Si), and Tungsten (W). Hence, molecular modeling of different compounds incorporated with three different force fields was evaluated in this work. In this study, we found that the In structure has more energy reduction, of 22.673 kJ mol−1 in UFF, when compared with the other two force fields. Thus, In has higher potential with more stability.

The applicability of the three steps systematic parametrization procedure (3SSPP) to develop a force field for primary amines was evaluated in the present work. Previous simulations of primary amines show that current force fields (FF) can underestimate some experimental values under room conditions. Therefore, we propose a new set of parameters, for an united atom (UA) model, that can be used for short and long amines which predict correctly thermodynamic and dynamical properties. Following the 3SSPP methodology, the partial charges are chosen to match the experimental dielectric constant whereas the Lennard-Jones (LJ) parameters, ε and σ, are fitted to reproduce the surface tension at the vapor-liquid interface and the liquid density, respectively. Simulations were initially conducted for the propylamine molecule by introducing three different types of carbon atoms, C_α and C_β, with electric charges, and C_n, without charge. Then, modifying the charges of the carbons and using the transferable LJ parameters, the new set of constants for long amines were found. The results show good agreement for the experimental dielectric constant and mass density with a percentage error less than 1% surface tension the error is up to 4% ethylamine, the new charges were obtained from a fitting function calculated from the long amines results. For these molecules, the values of the dielectric constant and the surface tension present errors of the order of 10% with the experimental data. Miscibility of the amines was also tested with the new parameters and the results show reasonable agreement with experiments.

Thermal conductivities of polybutadiene crosslinked with sulfur as a function of the heat flux autocorrelation function by using an equilibrium molecular dynamic (EMD) simulation were investigated. The Green–Kubo method was used to calculate thermal conductivities. All simulations were performed by applying the LAMMPS software (version 3 Mar 2020) package. The united-atom force field (OPLS-UA) from the Moltemplate software (version 2.20.3) was applied in the simulations. The influence of uniform and random distributions of sulfur in polybutadiene on the final value of thermal conductivities was studied by polymeric model structures with similar and variable degrees of crosslinking. The results showed that for identical degrees of crosslinking, the distribution of crosslinkers in the polymeric model structures significantly influenced the final value of thermal conductivity. Moreover, the influence of the crosslinking degree on the final value of thermal conductivity was studied by considering polymeric model structures with different degrees of crosslinking. The results demonstrate that by having a random distribution of sulfur, the thermal conductivity will be enhanced. However, by increasing the degree of crosslinking to the higher percentage in random crosslinked model structures, the value of thermal conductivity drops significantly due to possible higher crystallization of the model structures, which decrease the degree of freedom for phonon contributions.

Bonded (or valence) interactions, which directly determine the local structures of the molecules, are fundamental parts of molecular mechanics force fields (FFs). Most popular classical FFs adopt the simple harmonic models for bond stretching and angle bending and ignore cross-coupling effects among the valence terms. This may lead to less accurate vibrational properties and configurations in molecular dynamics (MD) simulations. AMOEBA models utilize an MM3(MM4)-style bonded interaction model, in which the vibrational anharmonicity, the coupling effects among different energy terms, and the out-of-plane bending for sp2-hybridized atoms are considered. In this work, we report the development of bonded interaction parameters for a wide range of chemistry based on quantum mechanics (QM). About 270 atomic types defined by SMARTS strings were used to model the valence interactions. Our results indicate that the resulting valence parameters produce accurate vibrational frequencies (RMSD from QM is less than ∼36.6cm−1) over a large set of molecules with diverse functional groups (445 molecules). By contrast, the harmonic models usually give an RMS error greater than 60cm−1. Meanwhile, this model accurately reflects the potential energy surface of the out-of-plane bending. Our model can generally be applied to the AMOEBA family and any MM3(MM4)-based molecular mechanics FFs. Bonded interactions are fundamental ingredients of molecular mechanics force fields because they directly determine the local structure of a molecule. In this work, we parametrize the advanced bonded energy functionals that consider the vibrational anharmonicity, the coupling effects, and the out-of-plane bending for sp2-hybridized atoms. It is expected that these models can describe the spectroscopic properties and overall structures of a molecule more accurately when they are used with polarizable AMOEBA-based force fields. [ABSTRACT FROM AUTHOR]

A new force field for 1-propanol, in the united and all atom models, has been obtained by combining two different empirical methodologies. The first was developed by scaling atom charges and Lennard-Jones parameters to fit the dielectric constant, surface tension, and density; this methodology is named three steps systematic parameterization procedure (3SSPP), as reported by Pérez de la Luz et al. (J Chem Theory Comput 14:5949–5958, 2018). The second methodology consists of moving these parameters and together with the bond distance to obtain the liquid-vapor phase diagram of the CO2 molecule as discussed by Harris and Yung (J Phys Chem 99:12021–12024, 1995). The last methodology is used to obtain the self-diffusion coefficient, which was not consider in the 3SSPP. The 3SSPP/bond methodology is the 3SSPP plus the bond distance scaling. With this new methodology, the experimental density, dielectric constant, surface tension, and self-diffusion coefficient at ambient temperature could be achieved. Furthermore, we show the temperature dependence of the aforementioned properties. The static structure factors are in accordance with the experimental spectrum. Solubility is increased to the experimental value for the united atom (UA) model after applying this methodology and for all atom (AA) scheme, the experimental solubility value is maintained. [ABSTRACT FROM AUTHOR]

Molecular mechanics force field calculations have historically shown significant limitations in modeling the energetic and conformational interconversions of highly substituted furanose rings. This is primarily due to the gauche effect that is not easily captured using pairwise energy potentials. In this study, we present a refinement to the set of torsional parameters in the General Amber Force Field (gaff) used to calculate the potential energy of mono, di-, and gem-fluorinated nucleosides. The parameters were optimized to reproduce the pseudorotation phase angle and relative energies of a diverse set of mono- and difluoro substituted furanose ring systems using quantum mechanics umbrella sampling techniques available in the IpolQ engine in the Amber suite of programs. The parameters were developed to be internally consistent with the gaff force field and the TIP3P water model. The new set of angle and dihedral parameters and partial charges were validated by comparing the calculated phase angle probability to those obtained from experimental nuclear magnetic resonance experiments. [ABSTRACT FROM AUTHOR]

Molecular dynamics (MD) simulation has been widely used in the field of biomedicine to study the conformational transition of proteins caused by mutation or ligand binding/unbinding. It provides some perspectives those are difficult to find in traditional biochemical or pathological experiments, for example, detailed effects of mutations on protein structure and protein–protein/ligand interaction at the atomic level. In this review, a broad overview on conformation changes and drug discovery by MD simulation is given. We first discuss the preparation of protein structure for MD simulation, which is a key step that determines the accuracy of the simulation. Then, we summarize the applications of commonly used force fields and MD simulations in scientific research. Finally, enhanced sampling methods and common applications of these methods are introduced. In brief, MD simulation is a powerful tool and it can be used to guide experimental study. The combination of MD simulation and experimental techniques is an a priori means to solve the biomedical problems and give a deep understanding on the relationship between protein structure and function. [ABSTRACT FROM AUTHOR]

We provide tailor-made GAFF2-type interaction potentials for modeling ammonium and amide ions in ammonia. Based on harmonic approximation of intra-molecular bond stretching and bending, our force fields nicely reproduce the vibrational modes of NH4+ and NH2−, respectively. Moreover, quantum calculations of pair-wise NH4+/NH2−–NH3 interactions were used for inter-molecular force field parameterization, while (NH3)n, [(NH4)(NH3)n]+, and [(NH2)(NH3)n]− complexes with n > 2, respectively, were reserved for benchmarking in terms of both structure and formation energy. Despite the limited reliability of molecular mechanics models for describing dimer complexes (n = 1), we find that GAFF2 reasonably reproduces [(NH4)(NH3)n]+ species for n = 2–4. For the assessment of [(NH2)(NH3)n]− complexes with n = 2–4, we however suggest the introduction of specific van der Waals parameters for amide-ammonia interactions. The application of the (extended) GAFF2 models is demonstrated for the study of ammonium and amide solvation in liquid ammonia at 240 K and 1 atm, respectively. On this basis, we suggest the applicability of our model for both gas phase and liquid states of ammonia. [ABSTRACT FROM AUTHOR]

Molecular dynamics (MD) simulation is a powerful tool for exploring the conformational energy landscape of proteins, and the reliability of MD results is crucially dependent on the underlying force field (FF). An accurate FF capable of producing balanced distributions of diverse conformations at multiple levels has been a long‐sought goal. Towards this, several decades of joint efforts have been made to address FF deficiencies, manifested by conformational biases at different levels (local conformations, secondary structures, and global extendedness of polypeptide chain). We first present the major FF biases, then review the strategies to address them separately. Specifically, both nonresidue‐specific and residue‐specific strategies for torsional parameter optimization have been applied to achieve local conformation and secondary structure balances. Significant improvements can be gained with residue‐specific torsional parameters especially when explicit dihedral couplings are considered. Further, the additional balance between protein–protein and protein–water interactions has been optimized via multiple ways to reproduce the global extendedness of polypeptide chains, especially for unfolded or disordered proteins. This review aims to summarize the most valuable experience and lessons gained from the past, which, we hope, can facilitate further improvements of both classical FFs and more sophisticated models such as polarizable FFs. This article is categorized under:Molecular and Statistical Mechanics > Molecular Dynamics and Monte‐Carlo MethodsMolecular and Statistical Mechanics > Molecular Mechanics [ABSTRACT FROM AUTHOR]

The calibration of torsional interaction terms by fitting relative gas‐phase conformational energies against their quantum‐mechanical values is a common procedure in force‐field development. However, much less attention has been paid to the optimization of third‐neighbor nonbonded interaction parameters, despite their strong coupling with the torsions. This article introduces an algorithm termed LLS‐SC, aimed at simultaneously parametrizing torsional and third‐neighbor interaction terms based on relative conformational energies. It relies on a self‐consistent (SC) procedure where each iteration involves a linear least‐squares (LLS) regression followed by a geometry optimization of the reference structures. As a proof‐of‐principle, this method is applied to obtain torsional and third‐neighbor interaction parameters for aliphatic chains in the context of the GROMOS 53A6 united‐atom force field. The optimized parameter set is compared to the original one, which has been fitted manually against thermodynamic properties for small linear alkanes. The LLS‐SC implementation is freely available under http://github.com/mssm-labmmol/profiler. [ABSTRACT FROM AUTHOR]

Discovery of natural antioxidants has been carried out for decades relying mainly on experimental approaches that are commonly associated with time and cost demanding biochemical assays. The maturation of quantitative structure activity relationship (QSAR) modelling has provided an alternative approach for searching and designing antioxidant compounds with alleviated costs. As a contribution to this approach, this work aimed to establish a fragment‐based 3D‐QSAR procedure to discover and design potential antioxidants based on tryptophyllin L structures isolated from the red tree frog Litoria rubella. A force field and a Gaussian 3D‐QSAR model were built to screen for potential antioxidants from tripeptide fragments covering all sequences of tryptophyllin L database. Among those, PWY(NH2) corresponding tryptophyllin L 4.1 was predicted to have the highest 2,2′‐azino‐bis(3‐ethylbenzothiazoline‐6‐sulphonic acid) radical cation (ABTS+·) scavenging capability. Two newly designed peptides PYW and PYW(NH2) together with PWY(NH2), tryptophyllin L 4.1, and the reference peptide PWY were synthesized and subjected to two antioxidant assays including ABTS scavenging and ferric reducing antioxidant power assays. Although the experimental TEAC values of the five peptides were roughly similar to those from predictions, the activity order was not in agreement with the predictions. The dissimilarities were accounted by the difference in the experimental procedures, the deviation of modelling regression, and the synergetic effect of structural and experimental features. The ABTS radical scavenging assays revealed that all the tested peptides were strong ABTS+· scavengers with the antioxidant capabilities approximately twice as high as trolox and higher than glutathione. The ferric reducing activities of the peptides were, on the other hand, much weaker than that of trolox suggesting different antioxidant mechanisms inserted by trolox and the peptides. This work was a demonstration that 3D‐QSAR methods can be employed in conjunction with experimental methods to effectively detect and design antioxidant peptides. [ABSTRACT FROM AUTHOR]

Hexagonal boron nitride (hBN), a promising 2D nanomaterial, has potential applications in desalination and osmotic energy harvesting. In all these applications, surface roughness significantly impacts fluid flow in nanomaterial, but its precise effect remains unclear. This creates a knowledge gap in understanding how surface roughness influences water flow at the water-hBN interface, which hinders the development of accurate molecular dynamics (MD) simulations. Here, we address this gap by employing density functional theory (DFT) to calculate atomic charges on rough hBN surfaces. These charges are incorporated into MD simulations, revealing a strong influence on the water-hBN interface. This combined approach accurately predicts experimental water slip length. We further quantify the water flow behavior on hBN using established force fields. Incorporating surface roughness into the model yields results in close agreement with the experimental slip length of ∼ 1 nm for water using FF-2 force fields, validating the simulation approach. Our findings highlight the importance of incorporating realistic surface roughness and force field models in MD simulations of water-nanomaterial interfaces. This work underscores the critical role of accurate 2D material models for understanding fluid flow in nanofluidic applications. [Display omitted] • Surface roughness accurately predicts water slip length (∼ 1 nm) on hBN. • Surface roughness increases friction and reduces slip length on hBN. • Potential to tailor slip length by manipulating roughness. • Close agreement with the experimental slip length of ∼ 1 nm using FF-2 force fields. • Study enhances understanding of nanoscale friction at water-hBN interfaces. [ABSTRACT FROM AUTHOR]

We re-investigate the image charge model of Iori and Corni (Iori and Corni, J. Comput. Chem. 2008, 29, 1656). We find that a simple symmetrization of their model allows to obtain quantitatively correct results for the electrostatic interaction of a water molecule with a metallic surface. This symmetrization reduces the magnitude of the electrostatic interaction to less than 10% of the total interaction energy. © 2017 Wiley Periodicals, Inc.

Intrinsically disordered proteins are proteins which lack of specific tertiary structure and unable to fold spontaneously without the partner binding. These intrinsically disordered proteins are found to associate with various diseases, such as diabetes, cancer, and neurodegenerative diseases. However, current widely used force fields, such as ff99SB, ff14SB, OPLS/AA, and Charmm27, are insufficient in sampling the conformational characters of intrinsically disordered proteins. In this study, the CMAP method was used to correct the φ/ψ distributions of disorder-promoting amino acids. The simulation results show that the force filed parameters (ff14IDPs) can improve the φ/ψ distributions of the disorder-promoting amino acids, with RMSD less than 0.10% relative to the benchmark data of intrinsically disordered proteins. Further test suggests that the calculated secondary chemical shifts under ff14IDPs are in quantitative agreement with the data of NMR experiment for five tested systems. In addition, the simulation results show that ff14IDPs can still be used to model structural proteins, such as tested lysozyme and ubiquitin, with better performance in coil regions than the original general Amber force field ff14SB. These findings confirm that the newly developed Amber ff14IDPs is a robust model for improving the conformation sampling of intrinsically disordered proteins.

The absolute binding free energies and binding enthalpies of twelve host-guest systems in the SAMPL5 blind challenge were computed using our attach-pull-release (APR) approach. This method has previously shown good correlations between experimental and calculated binding data in retrospective studies of cucurbit[7]uril (CB7) and β-cyclodextrin (βCD) systems. In the present work, the computed binding free energies for host octa acid (OA or OAH) and tetra-endo-methyl octa-acid (TEMOA or OAMe) with guests are in good agreement with prospective experimental data, with a coefficient of determination (R2) of 0.8 and root-mean-squared error of 1.7 kcal/mol using the TIP3P water model. The binding enthalpy calculations achieve moderate correlations, with R2 of 0.5 and RMSE of 2.5 kcal/mol, for TIP3P water. Calculations using the newly developed OPC water model also show good performance. Furthermore, the present calculations semi-quantitatively capture the experimental trend of enthalpy-entropy compensation observed, and successfully predict guests with the strongest and weakest binding affinity. The most populated binding poses of all twelve systems, based on clustering analysis of 750 ns molecular dynamics (MD) trajectories, were extracted and analyzed. Computational methods using MD simulations and explicit solvent models in a rigorous statistical thermodynamic framework, like APR, can generate reasonable predictions of binding thermodynamics. Especially with continuing improvement in simulation force fields, such methods hold the promise of making substantial contributions to hit identification and lead optimization in the drug discovery process.

[Display omitted] • All-atom simulations are impractical for bitumen microscale phenomena; we introduce a United Atom model alternative. • The force field surpasses other UA models when modelling large molecules, mirroring geometric, thermodynamic, and kinetic properties. • The force field features 17 bead types and 287 potentials, covering a variety of molecules for bitumen modelling. • The force field achieves a 100-fold increase in performance in MD simulations of bitumen. • The force field allows microsecond, micrometer MD simulations for bitumen SARA phase separation exploration. This paper presents a United Atom (UA) force field for simulating hydrocarbon molecules in bituminous materials, integrating explicit hydrogens into beads with their parent atom. This method simplifies all-atom molecular models, significantly accelerating Molecular Dynamics (MD) simulations of bitumen by 10 to 100 times. Key advantages include halving the particle count, eliminating complex hydrogen interactions, and decreasing the degrees of freedom of the molecules. Developed by mapping forces from an all-atom model to the centers of mass of UA model beads, the force field ensures accurate replication of energies, forces, and molecular conformations, mirroring properties like pressure and density. It features 17 bead types and 287 interaction types, encompassing various hydrocarbon molecules. The UA force field's stability, surpassing all-atom models, is a notable achievement. This stability, stemming from smoother potential energy surfaces, leads to consistent property measurements and improved stress tensor accuracy. It enables the extension of MD simulations to larger spatiotemporal scales, crucial for understanding complex phenomena such as phase separation in bituminous materials. This foundational work sets the stage for future developments, including refining parameters and introducing new bead types, to enhance the modeling capabilities of the force field, thereby advancing the application and understanding of bituminous materials. [ABSTRACT FROM AUTHOR]

This paper reviews molecular dynamics (MD) concepts on heat transfer analysis of supercritical CO2, and highlights the major parameters that can affect the accuracy of respective thermal coefficients. Subsequently, the prime aspects of construction, transfer identification, and thermal performance are organized according to their challenges and prospective solutions associated with the mutability of supercritical CO2 properties. Likewise, the characteristics of bound force field schemes and thermal relaxation approaches are discussed on a case-by-case basis. Both convective and diffusive states of trans- and supercritical CO2 are debated, given their magnitude effects on molecular interactions. Following the scarcity of literature on similar enquiries, this paper recommended a future series of studies on molecular dynamics models in a large region of supercriticality and phase-interactions for coupled heat and mass transfer systems. This review recognizes that the foremost undertaking is to ascertain the thermo-hydraulic identity of supercritical CO2 for process feasibility of developed technology.

Host–guest binding, despite the relatively simple structural and chemical features of individual components, still poses a challenge in computational modelling. The extreme underperformance of standard end-point methods in host–guest binding makes them practically useless. In the current work, we explore a potentially promising modification of the three-trajectory realization. The alteration couples the binding-induced structural reorganization into free energy estimation and suffers from dramatic fluctuations in internal energies in protein–ligand situations. Fortunately, the relatively small size of host–guest systems minimizes the magnitude of internal fluctuations and makes the three-trajectory realization practically suitable. Due to the incorporation of intra-molecular interactions in free energy estimation, a strong dependence on the force field parameters could be incurred. Thus, a term-specific investigation of transferable GAFF derivatives is presented, and noticeable differences in many aspects are identified between commonly applied GAFF and GAFF2. These force-field differences lead to different dynamic behaviors of the macrocyclic host, which ultimately would influence the end-point sampling and binding thermodynamics. Therefore, the three-trajectory end-point free energy calculations are performed with both GAFF versions. Additionally, due to the noticeable differences between host dynamics under GAFF and GAFF2, we add additional benchmarks of the single-trajectory end-point calculations. When only the ranks of binding affinities are pursued, the three-trajectory realization performs very well, comparable to and even better than the regressed PBSA_E scoring function and the dielectric constant-variable regime. With the GAFF parameter set, the TIP3P water in explicit solvent sampling and either PB or GB implicit solvent model in free energy estimation, the predictive power of the three-trajectory realization in ranking calculations surpasses all existing end-point methods on this dataset. We further combine the three-trajectory realization with another promising modified end-point regime of varying the interior dielectric constant. The combined regime does not incur sizable improvements for ranks and deviations from experiment exhibit non-monotonic variations.

Boron containing compounds (BCCs) aroused increasing interest in the scientific community due to their wide application as drugs in various fields. In order to design new compounds hopefully endowed with pharmacological activity and also investigate their conformational behavior, the support of computational studies is crucial. Nevertheless, the suitable molecular mechanics parameterization and the force fields needed to perform these simulations are not completely available for this class of molecules. In this paper, Amber force field parameters for phenyl-, benzyl-, benzylamino-, and methylamino-boronates, a group of boron-containing compounds involved in different branches of the medicinal chemistry, were created. The robustness of the obtained data was confirmed through molecular dynamics simulations on ligand/β-lactamases covalent complexes. The ligand torsional angles, populated over the trajectory frames, were confirmed by values found in the ligand geometries, located through optimizations at the DFT/B3LYP/6-31g(d) level, using water as a solvent. In summary, this study successfully provided a library of parameters, opening the possibility to perform molecular dynamics simulations of this class of boron-containing compounds.

Predicting energies and forces using machine learning force field (MLFF) depends on accurate descriptions (features) of chemical environment. Despite the numerous features proposed, there is a lack of controlled comparison among them for their universality and accuracy. In this work, we compared several commonly used feature types for their ability to describe physical systems. These different feature types include cosine feature, Gaussian feature, moment tensor potential (MTP) feature, spectral neighbor analysis potential feature, simplified smooth deep potential with Chebyshev polynomials feature and Gaussian polynomials feature, and atomic cluster expansion feature. We evaluated the training root mean square error (RMSE) for the atomic group energy, total energy, and force using linear regression model regarding to the density functional theory results. We applied these MLFF models to an amorphous sulfur system and carbon systems, and the fitting results show that MTP feature can yield the smallest RMSE results compared with other feature types for either sulfur system or carbon system in the disordered atomic configurations. Moreover, as an extending test of other systems, the MTP feature combined with linear regression model can also reproduce similar quantities along the ab initio molecular dynamics trajectory as represented by Cu systems. Our results are helpful in selecting the proper features for the MLFF development.

Hydration free energies of small molecules are commonly used as benchmarks for solvation models. However, errors in predicting hydration free energies are partially due to the force fields used and not just the solvation model. To address this, we have used the 3D reference interaction site model (3D-RISM) of molecular solvation and existing benchmark explicit solvent calculations with a simple element count correction (ECC) to identify problems with the non-bond parameters in the general AMBER force field (GAFF). 3D-RISM was used to calculate hydration free energies of all 642 molecules in the FreeSolv database, and a partial molar volume correction (PMVC), ECC, and their combination (PMVECC) were applied to the results. The PMVECC produced a mean unsigned error of 1.01±0.04kcal/mol and root mean squared error of 1.44±0.07kcal/mol, better than the benchmark explicit solvent calculations from FreeSolv, and required less than 15 s of computing time per molecule on a single CPU core. Importantly, parameters for PMVECC showed systematic errors for molecules containing Cl, Br, I, and P. Applying ECC to the explicit solvent hydration free energies found the same systematic errors. The results strongly suggest that some small adjustments to the Lennard–Jones parameters for GAFF will lead to improved hydration free energy calculations for all solvent models.