Predicting Hydrophobicity by Learning Spatiotemporal Features of Interfacial Water Structure: Combining Molecular Dynamics Simulations with Convolutional Neural Networks

Converting Nanotoxicity Data to Information Using Artificial Intelligence and Simulation

Decades of nanotoxicology research have generated extensive and diverse data sets. However, data is not equal to information. The question is how to extract critical information buried in vast data streams. Here we show that artificial intelligence (AI) and molecular simulation play key roles in transforming nanotoxicity data into critical information, i.e., constructing the quantitative nanostructure (physicochemical properties)-toxicity relationships, and elucidating the toxicity-related molecular mechanisms. For AI and molecular simulation to realize their full impacts in this mission, several obstacles must be overcome. These include the paucity of high-quality nanomaterials (NMs) and standardized nanotoxicity data, the lack of model-friendly databases, the scarcity of specific and universal nanodescriptors, and the inability to simulate NMs at realistic spatial and temporal scales. This review provides a comprehensive and representative, but not exhaustive, summary of the current capability gaps and tools required to fill these formidable gaps. Specifically, we discuss the applications of AI and molecular simulation, which can address the large-scale data challenge for nanotoxicology research. The need for model-friendly nanotoxicity databases, powerful nanodescriptors, new modeling approaches, molecular mechanism analysis, and design of the next-generation NMs are also critically discussed. Finally, we provide a perspective on future trends and challenges.

Oxide- and Silicate-Water Interfaces and Their Roles in Technology and the Environment

Interfacial reactions drive all elemental cycling on Earth and play pivotal roles in human activities such as agriculture, water purification, energy production and storage, environmental contaminant remediation, and nuclear waste repository management. The onset of the 21st century marked the beginning of a more detailed understanding of mineral aqueous interfaces enabled by advances in techniques that use tunable high-flux focused ultrafast laser and X-ray sources to provide near-atomic measurement resolution, as well as by nanofabrication approaches that enable transmission electron microscopy in a liquid cell. This leap into atomic- and nanometer-scale measurements has uncovered scale-dependent phenomena whose reaction thermodynamics, kinetics, and pathways deviate from previous observations made on larger systems. A second key advance is new experimental evidence for what scientists hypothesized but could not test previously, namely, interfacial chemical reactions are frequently driven by "anomalies" or "non-idealities" such as defects, nanoconfinement, and other nontypical chemical structures. Third, progress in computational chemistry has yielded new insights that allow a move beyond simple schematics, leading to a molecular model of these complex interfaces. In combination with surface-sensitive measurements, we have gained knowledge of the interfacial structure and dynamics, including the underlying solid surface and the immediately adjacent water and aqueous ions, enabling a better definition of what constitutes the oxide- and silicate-water interfaces. This critical review discusses how science progresses from understanding ideal solid-water interfaces to more realistic systems, focusing on accomplishments in the last 20 years and identifying challenges and future opportunities for the community to address. We anticipate that the next 20 years will focus on understanding and predicting dynamic transient and reactive structures over greater spatial and temporal ranges as well as systems of greater structural and chemical complexity. Closer collaborations of theoretical and experimental experts across disciplines will continue to be critical to achieving this great aspiration.

Molecular Dynamics Investigation of Nanoscale Hydrophobicity of Polymer Surfaces: What Makes Water Wet?

The wettability of a polymer surface─related to its hydrophobicity or tendency to repel water─can be crucial for determining its utility, such as for a coating or a purification membrane. While wettability is commonly associated with the macroscopic measurement of a contact angle between surface, water, and air, the molecular physics that underlie these macroscopic observations are not fully known, and anticipating the relative behavior of different polymers is challenging. To address this gap in molecular-level understanding, we use molecular dynamics simulations to investigate and contrast interactions of water with six chemically distinct polymers: polytetrafluoroethylene, polyethylene, polyvinyl chloride, poly(methyl methacrylate), Nylon-66, and poly(vinyl alcohol). We show that several prospective quantitative metrics for hydrophobicity agree well with experimental contact angles. Moreover, the behavior of water in proximity to these polymer surfaces can be distinguished with analysis of interfacial water dynamics, extent of hydrogen bonding, and molecular orientation─even when macroscopic measures of hydrophobicity are similar. The predominant factor dictating wettability is found to be the extent of hydrogen bonding between polymer and water, but the precise manifestation of hydrogen bonding and its impact on surface water structure varies. In the absence of hydrogen bonding, other molecular interactions and polymer mechanics control hydrophobic ordering. These results provide new insights into how polymer chemistry specifically impacts water-polymer interactions and translates to surface hydrophobicity. Such factors may facilitate the design or processing of polymer surfaces to achieve targeted wetting behavior, and presented analyses can be useful in studying the interfacial physics of other systems.

Topological Analysis of Molecular Dynamics Simulations using the Euler Characteristic

Molecular dynamics (MD) simulations are used in diverse scientific and engineering fields such as drug discovery, materials design, separations, biological systems, and reaction engineering. These simulations generate highly complex data sets that capture the 3D spatial positions, dynamics, and interactions of thousands of molecules. Analyzing MD data sets is key for understanding and predicting emergent phenomena and in identifying key drivers and tuning design knobs of such phenomena. In this work, we show that the Euler characteristic (EC) provides an effective topological descriptor that facilitates MD analysis. The EC is a versatile, low-dimensional, and easy-to-interpret descriptor that can be used to reduce, analyze, and quantify complex data objects that are represented as graphs/networks, manifolds/functions, and point clouds. Specifically, we show that the EC is an informative descriptor that can be used for machine learning and data analysis tasks such as classification, visualization, and regression. We demonstrate the benefits of the proposed approach through case studies that aim to understand and predict the hydrophobicity of self-assembled monolayers and the reactivity of complex solvent environments.

Structural features of interfacial water predict the hydrophobicity of chemically heterogeneous surfaces

The hydrophobicity of an interface determines the magnitude of hydrophobic interactions that drive numerous biological and industrial processes. Chemically heterogeneous interfaces are abundant in these contexts; examples include the surfaces of proteins, functionalized nanomaterials, and polymeric materials. While the hydrophobicity of nonpolar solutes can be predicted and related to the structure of interfacial water molecules, predicting the hydrophobicity of chemically heterogeneous interfaces remains a challenge because of the complex, non-additive contributions to hydrophobicity that depend on the chemical identity and nanoscale spatial arrangements of polar and nonpolar groups. In this work, we utilize atomistic molecular dynamics simulations in conjunction with enhanced sampling and data-centric analysis techniques to quantitatively relate changes in interfacial water structure to the hydration free energy (a thermodynamically well-defined descriptor of hydrophobicity) of chemically heterogeneous interfaces. We analyze a large data set of 58 self-assembled monolayers (SAMs) composed of ligands with nonpolar and polar end groups of different chemical identity (amine, amide, and hydroxyl) in five mole fractions, two spatial patterns, and with scaled partial charges. We find that only five features of interfacial water structure are required to accurately predict hydration free energies. Examination of these features reveals mechanistic insights into the interfacial hydrogen bonding behaviors that distinguish different surface compositions and patterns. This analysis also identifies the probability of highly coordinated water structures as a unique signature of hydrophobicity. These insights provide a physical basis to understand the hydrophobicity of chemically heterogeneous interfaces and connect hydrophobicity to experimentally accessible perturbations of interfacial water structure.

Predicting the Physicochemical Properties and Biological Activities of Monolayer-Protected Gold Nanoparticles Using Simulation-Derived Descriptors

Gold nanoparticles are versatile materials for biological applications because their properties can be modulated by assembling ligands on their surface to form monolayers. However, the physicochemical properties and behaviors of monolayer-protected nanoparticles in biological environments are difficult to anticipate because they emerge from the interplay of ligand-ligand and ligand-solvent interactions that cannot be readily inferred from ligand chemical structure alone. In this work, we demonstrate that quantitative nanostructure-activity relationship (QNAR) models can employ descriptors calculated from molecular dynamics simulations to predict nanoparticle properties and cellular uptake. We performed atomistic molecular dynamics simulations of 154 monolayer-protected gold nanoparticles and calculated a small library of simulation-derived descriptors that capture nanoparticle structural and chemical properties in aqueous solution. We then parametrized QNAR models using interpretable regression algorithms to predict experimental measurements of nanoparticle octanol-water partition coefficients, zeta potentials, and cellular uptake obtained from a curated database. These models reveal that simulation-derived descriptors can accurately predict experimental trends and provide physical insight into what descriptors are most important for obtaining desired nanoparticle properties or behaviors in biological environments. Finally, we demonstrate model generalizability by predicting cell uptake trends for 12 nanoparticles not included in the original data set. These results demonstrate that QNAR models parametrized with simulation-derived descriptors are accurate, generalizable computational tools that could be used to guide the design of monolayer-protected gold nanoparticles for biological applications without laborious trial-and-error experimentation.

Identifying nonadditive contributions to the hydrophobicity of chemically heterogeneous surfaces via dual-loop active learning

Hydrophobic interactions drive numerous biological and synthetic processes. The materials used in these processes often possess chemically heterogeneous surfaces that are characterized by diverse chemical groups positioned in close proximity at the nanoscale; examples include functionalized nanomaterials and biomolecules, such as proteins and peptides. Nonadditive contributions to the hydrophobicity of such surfaces depend on the chemical identities and spatial patterns of polar and nonpolar groups in ways that remain poorly understood. Here, we develop a dual-loop active learning framework that combines a fast reduced-accuracy method (a convolutional neural network) with a slow higher-accuracy method (molecular dynamics simulations with enhanced sampling) to efficiently predict the hydration free energy, a thermodynamic descriptor of hydrophobicity, for nearly 200 000 chemically heterogeneous self-assembled monolayers (SAMs). Analysis of this dataset reveals that SAMs with distinct polar groups exhibit substantial variations in hydrophobicity as a function of their composition and patterning, but the clustering of nonpolar groups is a common signature of highly hydrophobic patterns. Further molecular dynamics analysis relates such clustering to the perturbation of interfacial water structure. These results provide new insight into the influence of chemical heterogeneity on hydrophobicity via quantitative analysis of a large set of surfaces, enabled by the active learning approach.

Hydration Dynamics in Biological Membranes: Emerging Applications of Terahertz Spectroscopy

Water drives the spontaneous self-assembly of lipids and proteins into quasi two-dimensional biological membranes that act as catalytic scaffolds for numerous processes central to life. However, the functional relevance of hydration in membrane biology is only beginning to be addressed, predominantly because of challenges associated with direct measurements of hydration microstructure and dynamics in a biological milieu. Our recent work on the novel interplay of membrane electrostatics and crowding in shaping membrane hydration dynamics utilizing terahertz (THz) spectroscopy represents an important step in this context. In this Perspective, we provide a glimpse into the ever-broadening functional landscape of hydration dynamics in biological membranes in the backdrop of the unique physical chemistry of water molecules. We further highlight the immense (and largely untapped) potential of the THz toolbox in addressing contemporary problems in membrane biology, while emphasizing the adaptability of the analytical framework reported recently by us to such studies.

AI-driven prediction of SARS-CoV-2 variant binding trends from atomistic simulations

We present a novel technique to predict binding affinity trends between two molecules from atomistic molecular dynamics simulations. The technique uses a neural network algorithm applied to a series of images encoding the distance between two molecules in time. We demonstrate that our algorithm is capable of separating with high accuracy non-hydrophobic mutations with low binding affinity from those with high binding affinity. Moreover, we show high accuracy in prediction using a small subset of the simulation, therefore requiring a much shorter simulation time. We apply our algorithm to the binding between several variants of the SARS-CoV-2 spike protein and the human receptor ACE2.

Geometric Feature Extraction of Point Cloud of Chemical Reactor Based on Dynamic Graph Convolution Neural Network

Geometric features are an important factor for the classification of drugs and other transport objects in chemical reactors. The moving speed of drugs and other transport objects in chemical reactors is fast, and it is difficult to obtain their features by imaging and other methods. In order to avoid the mistaken and missed distribution of drugs and other objects, a method of extracting geometric features of the drug's point cloud in a chemical reactor based on a dynamic graph convolution neural network (DGCNN) is proposed. In this study, we first use MATLAB R2019a to add a random number of noise points in each point cloud file and label the point cloud. Second, k-nearest neighbor (KNN) is used to construct the adjacency relationship of all nodes, and the effect of DGCNN under different k values and the confusion matrix under the optimal k value are analyzed. Finally, we compare the effect of DGCNN with PointNet and PointNet++. The experimental results show that when k is 20, the accuracy, precision, recall, and F1 score of DGCNN are higher than those of other k values, while the training time is much shorter than that of k = 25, 30, and 35; in addition, the effect of DGCNN in extracting geometric features of the point cloud is better than that of PointNet and PointNet++. The results show that it is feasible to use DGCNN to analyze the geometric characteristics of drug point clouds in a chemical reactor. This study fills the gap of the end-to-end extraction method for a point cloud's corresponding geometric features without a data set. In addition, this study promotes the institutionalization, standardization, and intelligent design of safe production and management of drugs and other objects in the chemical reactor, and it has positive significance for the production cost and resource utilization of the whole pharmaceutical process. At the same time, it provides a new method for the intelligent processing of point cloud data.

The Interplay of Ligand Properties and Core Size Dictates the Hydrophobicity of Monolayer-Protected Gold Nanoparticles

The hydrophobicity of monolayer-protected gold nanoparticles is a crucial design parameter that influences self-assembly, preferential binding to proteins and membranes, and other nano-bio interactions. Predicting the effects of monolayer components on nanoparticle hydrophobicity is challenging due to the nonadditive, cooperative perturbations to interfacial water structure that dictate hydrophobicity at the nanoscale. In this work, we quantify nanoparticle hydrophobicity by using atomistic molecular dynamics simulations to calculate local hydration free energies at the nanoparticle-water interface. The simulations reveal that the hydrophobicity of large gold nanoparticles is determined primarily by ligand end group chemistry, as expected. However, for small gold nanoparticles, long alkanethiol ligands interact to form anisotropic bundles that lead to substantial spatial variations in hydrophobicity even for homogeneous monolayer compositions. We further show that nanoparticle hydrophobicity is modulated by changing the ligand structure, ligand chemistry, and gold core size, emphasizing that single-ligand properties alone are insufficient to characterize hydrophobicity. Finally, we illustrate that hydration free energy measurements correlate with the preferential binding of propane as a representative hydrophobic probe molecule. Together, these results show that both physical and chemical properties influence the hydrophobicity of small nanoparticles and must be considered together when predicting gold nanoparticle interactions with biomolecules.