Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems

Extraction of short-chain fatty acid ethyl Ester in Baijiu using covalent organic framework-based magnetic nanoparticles: Theoretical screening and experimental validation

Short-chain fatty acid ethyl esters (SFAEEs) are critical aroma compounds in Baijiu, and their wider concentration range can lead to differences in the quality grade of Baijiu. Efficiently designing an SFAEEs adsorbent before instrument analysis remains challenging. In this work, nine functionalized covalent organic frameworks (COFs) with different postmodification groups were designed for targeting SFAEEs. Based on interaction energy as the evaluation criterion, COFs modified with 5-Mercapto-1-methyltetrazole (MMTZ) had been identified through density functional theory screening. Using imine COFs and MMTZ, novel magnetic nanoparticles (Fe3O4@COFs@MMTZ) were prepared and used to develop the magnetic solid-phase extraction of SFAEEs from Baijiu. The adsorption mechanism of Fe3O4@COFs@MMTZ was analyzed using wave function analysis, revealing that adsorption occurred via vdW interaction, CH···π interaction, and hydrogen bonding. This study provides a new concept for the rapid detection of SFAEEs and theoretical support for the scientific construction of quality control during Baijiu production.

Advancing Multiscale Molecular Modeling with Machine Learning-Derived Electrostatics

We introduce an innovative machine learning (ML)-based framework for multiscale molecular modeling in which the ML subsystem is treated as an electrostatic entity interacting with its molecular mechanics (MM) environment through classical electrostatics. The integration of ML accuracy with multiscale modeling is accomplished by leveraging the capabilities of the ANI neural networks to predict geometry-dependent atomic partial charges at the minimal basis iterative stockholder (MBIS) level, going beyond static mechanical embedding. This ML/MM approach can closely approximate state-of-the-art multiscale quantum-classical (QM/MM) methods while significantly lowering computational requirements, thereby facilitating more efficient and precise simulations in computational chemistry. The method requires no additional training beyond the initial model setup and is integrated into Amber, one of the most widely used software suites for molecular modeling, ensuring accessibility to the broader community. We validate its performance across a variety of challenging applications, including the solvation structure, vibrational spectra, torsion free energy profiles, and protein-ligand interactions, achieving excellent agreement with QM/MM benchmarks. This framework not only advances the frontiers of multiscale modeling but also showcases the potential of machine learning to achieve quantum-level accuracy with exceptional efficiency for complex chemical systems.

Machine learning in molecular biophysics: Protein allostery, multi-level free energy simulations, and lipid phase transitions

Machine learning (ML) techniques have been making major impacts on all areas of science and engineering, including biophysics. In this review, we discuss several applications of ML to biophysical problems based on our recent research. The topics include the use of ML techniques to identify hotspot residues in allosteric proteins using deep mutational scanning data and to analyze how mutations of these hotspots perturb co-operativity in the framework of a statistical thermodynamic model, to improve the accuracy of free energy simulations by integrating data from different levels of potential energy functions, and to determine the phase transition temperature of lipid membranes. Through these examples, we illustrate the unique value of ML in extracting patterns or parameters from complex data sets, as well as the remaining limitations. By implementing the ML approaches in the context of physically motivated models or computational frameworks, we are able to gain a deeper mechanistic understanding or better convergence in numerical simulations. We conclude by briefly discussing how the introduced models can be further expanded to tackle more complex problems.

AtomDB: A Python Library and Database for Atomic and Promolecular Properties

AtomDB is a free and open-source Python library for accessing and manipulating neutral and charged atomic species and their promolecular properties. It serves as a computational toolset, operating on an accompanying "extended periodic table" database, with experimental and computational data covering atomic species with a wide range of charges and multiplicities. AtomDB includes facilities for computing promolecules: local promolecular properties, constructed from the corresponding atomic densities, and scalar promolecular properties, computed from the corresponding scalar atomic properties, both taking into account whether properties are extensive or intensive. AtomDB is designed to be easy to use, extend, and maintain: it follows best practices for modern software development, including comprehensive documentation, extensive testing, continuous integration/delivery protocols, and package management. This article is the official release note for the AtomDB library.

Pretraining graph transformers with atom-in-a-molecule quantum properties for improved ADMET modeling

We evaluate the impact of pretraining Graph Transformer architectures on atom-level quantum-mechanical features for the modeling of absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of drug-like compounds. We compare this pretraining strategy with two others: one based on molecular quantum properties (specifically the HOMO-LUMO gap) and one using a self-supervised atom masking technique. After fine-tuning on Therapeutic Data Commons ADMET datasets, we evaluate the performance improvement in the different models observing that models pretrained with atomic quantum mechanical properties produce in general better results. We then analyze the latent representations and observe that the supervised strategies preserve the pretraining information after fine-tuning and that different pretrainings produce different trends in latent expressivity across layers. Furthermore, we find that models pretrained on atomic quantum mechanical properties capture more low-frequency Laplacian eigenmodes of the input graph via the attention weights and produce better representations of atomic environments within the molecule. Application of the analysis to a much larger non-public dataset for microsomal clearance illustrates generalizability of the studied indicators. In this case the performances of the models are in accordance with the representation analysis and highlight, especially for the case of masking pretraining and atom-level quantum property pretraining, how model types with similar performance on public benchmarks can have different performances on large scale pharmaceutical data.Scientific contributionWe systematically compared three different data type/methodologies for pretraining molecular Graphormer with the purpose of modeling ADMET properties as downstream tasks. The learned representations from differently pretrained models were analyzed in addition to comparison of downstream task performances that have been typically reported in similar works. Such examination methodologies, including a newly introduced analysis of Graphormer's Attention Rollout Matrix, can guide pretraining strategy selection, as corroborated by a performance evaluation on a larger internal dataset.

Predicting the Mutagenic Activity of Nitroaromatics Using Conceptual Density Functional Theory Descriptors and Explainable No-Code Machine Learning Approaches

Nitroaromatic compounds (NAs) are widely used in industrial applications but pose significant genotoxic risks, necessitating accurate mutagenicity prediction for chemical safety assessments. This study integrates conceptual density functional theory (CDFT) descriptors with explainable no-code machine learning (ML) models to predict NA mutagenicity based on Ames test results. Following OECD QSAR guidelines, feature selection and model development were performed using decision-tree-based algorithms (Random Tree, JCHAID*, SPAARC) and multilayer perceptrons (MLPs). These models exhibited high predictive accuracy (internal: >80%, κ = 0.21-0.37; external: ∼90%, κ = 0.41-0.62) with strong interpretability. The study also explores the role of metabolic activation and aqueous-phase descriptors, evaluating a novel electronic analog to LogP (LogQP) to assess hydrophobicity-mutagenicity relationships. Results demonstrate that aqueous-phase electronic properties and electrophilicity descriptors outperform vacuum-based methods in mutagenicity prediction. The combination of CDFT descriptors with shallow ML models proves to be a robust, interpretable, and accessible framework for predictive toxicology. This approach enhances chemical risk assessment and bridges computational chemistry with toxicology for regulatory applications.

Structural Similarity, Activity, and Toxicity of Mycotoxins: Combining Insights from Unsupervised and Supervised Machine Learning Algorithms

A large number of mycotoxins and related fungal metabolites have not been assessed in terms of their toxicological impacts. Current methodologies often prioritize specific target families, neglecting the complexity and presence of co-occurring compounds. This work addresses a fundamental question: Can we assess molecular similarity and predict the toxicity of mycotoxins in silico using a defined set of molecular descriptors? We propose a rapid nontarget screening approach for multiple classes of mycotoxins, integrating both unsupervised and supervised machine learning models, alongside molecular and physicochemical descriptors to enhance the understanding of structural similarity, activity, and toxicity. Clustering analyses identify natural clusters corresponding to the known mycotoxin families, indicating that mycotoxins belonging to the same cluster share similar molecular properties. However, topological descriptors play a significant role in distinguishing between acutely toxic and nonacutely toxic compounds. Random forest (RF) and neural networks (NN), combined with molecular descriptors, contribute to improved knowledge and predictive capability regarding mycotoxin toxicity profiles. RF allows the prediction of toxicity using data reflecting mainly structural features and performs well in the presence of descriptors reflecting biological activity. NN models prove to be more sensitive to biological activity descriptors than RF. The use of descriptors encompassing structural complexity and diversity, chirality and symmetry, connectivity, atomic charge, and polarizability, together with descriptors representing lipophilicity, absorption, and permeation of molecules, is crucial for predicting toxicity, facilitating broader toxicological evaluations.

Neural Network Potential with Multiresolution Approach Enables Accurate Prediction of Reaction Free Energies in Solution

We present the design and implementation of a novel neural network potential (NNP) and its combination with an electrostatic embedding scheme, commonly used within the context of hybrid quantum-mechanical/molecular-mechanical (QM/MM) simulations. Substitution of a computationally expensive QM Hamiltonian by an NNP with the same accuracy largely reduces the computational cost and enables efficient sampling in prospective MD simulations, the main limitation faced by traditional QM/MM setups. The model relies on the recently introduced anisotropic message passing (AMP) formalism to compute atomic interactions and encode symmetries found in QM systems. AMP is shown to be highly efficient in terms of both data and computational costs and can be readily scaled to sample systems involving more than 350 solute and 40,000 solvent atoms for hundreds of nanoseconds using umbrella sampling. Most deviations of AMP predictions from the underlying DFT ground truth lie within chemical accuracy (4.184 kJ mol-1). The performance and broad applicability of our approach are showcased by calculating the free-energy surface of alanine dipeptide, the preferred ligation states of nickel phosphine complexes, and dissociation free energies of charged pyridine and quinoline dimers. Results with this ML/MM approach show excellent agreement with experimental data and reach chemical accuracy in most cases. In contrast, free energies calculated with static DFT calculations paired with implicit solvent models or QM/MM MD simulations using cheaper semiempirical methods show up to ten times higher deviation from the experimental ground truth and sometimes even fail to reproduce qualitative trends.

The Influence of Ionizing Radiation on Quantification for In Situ and Operando Liquid-Phase Electron Microscopy

The ionizing radiation harnessed in electron microscopes or synchrotrons enables unique insights into nanoscale dynamics. In liquid-phase transmission electron microscopy (LP-TEM), irradiating a liquid sample with electrons offers access to real space information at an unmatched combination of temporal and spatial resolution. However, employing ionizing radiation for imaging can alter the Gibbs free energy landscape during the experiment. This is mainly due to radiolysis and the corresponding shift in chemical potential; however, experiments can also be affected by irradiation-induced charging and heating. In this review, the state of the art in describing beam effects is summarized, theoretical and experimental assessment guidelines are provided, and strategies to obtain quantitative information under such conditions are discussed. While this review showcases these effects on LP-TEM, the concepts that are discussed here can also be applied to other types of ionizing radiation used to probe liquid samples, such as synchrotron X-rays.

Crash testing machine learning force fields for molecules, materials, and interfaces: model analysis in the TEA Challenge 2023

Atomistic simulations are routinely employed in academia and industry to study the behavior of molecules, materials, and their interfaces. Central to these simulations are force fields (FFs), whose development is challenged by intricate interatomic interactions at different spatio-temporal scales and the vast expanse of chemical space. Machine learning (ML) FFs, trained on quantum-mechanical energies and forces, have shown the capacity to achieve sub-kcal (mol-1 Å-1) accuracy while maintaining computational efficiency. The TEA Challenge 2023 rigorously evaluated commonly used MLFFs across diverse applications, highlighting their strengths and weaknesses. Participants trained their models using provided datasets, and the results were systematically analyzed to assess the ability of MLFFs to reproduce potential energy surfaces, handle incomplete reference data, manage multi-component systems, and model complex periodic structures. This publication describes the datasets, outlines the proposed challenges, and presents a detailed analysis of the accuracy, stability, and efficiency of the MACE, SO3krates, sGDML, SOAP/GAP, and FCHL19* architectures in molecular dynamics simulations. The models represent the MLFF developers who participated in the TEA Challenge 2023. All results presented correspond to the state of the ML architectures as of October 2023. A comprehensive analysis of the molecular dynamics results obtained with different MLFFs will be presented in the second part of this manuscript.

Development and validation of a support vector machine-based nomogram for diagnosis of obstetric antiphospholipid syndrome

BACKGROUND

Antiphospholipid Syndrome (APS) is a systemic autoimmune disorder characterized by arterial or venous thrombosis and/or pregnancy complications. This study aims to develop a diagnostic model for Obstetric APS (OAPS) using the Support Vector Machine (SVM) algorithm.

METHODS

Data were retrospectively collected from 102 patients with OAPS and 80 healthy controls (HC). Utilizing random sampling, patients were randomly allocated into a training set and a validation set. The training set comprised 72 OAPS patients and 52 HCs, while the validation set included 30 OAPS patients and 24 HCs. Univariate logistic regression analysis and the LASSO method were employed to screen feature variables. Subsequently, the selected feature variables were used to construct a diagnostic model based on the SVM algorithm, which was then validated within the training set.

RESULTS

An optimal subset comprising 12 clinical features was curated. This ensemble of clinical features exhibited formidable predictive efficacy within both the training and validation datasets, as evidenced by Area Under the Curve (AUC) values of 0.969 and 0.942, sensitivities of 0.875 and 0.867, and specificities of 0.929 and 0.875, respectively. Furthermore, the nomogram generated a Concordance Index (C-index) of 0.851 across the entire dataset. Decision curve analysis demonstrates that the combined nomogram and TAT nomogram offer greater net benefit compared to nomograms based on other individual clinical indicators within the dataset.

CONCLUSION

The SVM-based model can effectively diagnose patients with OAPS.

A review of large language models and autonomous agents in chemistry

Large language models (LLMs) have emerged as powerful tools in chemistry, significantly impacting molecule design, property prediction, and synthesis optimization. This review highlights LLM capabilities in these domains and their potential to accelerate scientific discovery through automation. We also review LLM-based autonomous agents: LLMs with a broader set of tools to interact with their surrounding environment. These agents perform diverse tasks such as paper scraping, interfacing with automated laboratories, and synthesis planning. As agents are an emerging topic, we extend the scope of our review of agents beyond chemistry and discuss across any scientific domains. This review covers the recent history, current capabilities, and design of LLMs and autonomous agents, addressing specific challenges, opportunities, and future directions in chemistry. Key challenges include data quality and integration, model interpretability, and the need for standard benchmarks, while future directions point towards more sophisticated multi-modal agents and enhanced collaboration between agents and experimental methods. Due to the quick pace of this field, a repository has been built to keep track of the latest studies: https://github.com/ur-whitelab/LLMs-in-science.

High-Throughput Combinatorial Metal Complex Synthesis

High-throughput combinatorial metal complex synthesis has emerged as a powerful tool for rapidly generating and screening diverse libraries of metal complexes, enabling accelerated discovery in fields such as catalysis, medicinal chemistry, and materials science. By systematically combining building blocks under mild and efficient conditions, researchers can explore broad chemical spaces, increasing the likelihood of identifying complexes with desired properties. This method streamlines hit identification and optimisation, especially when integrated with high-throughput screening and data-driven approaches like machine learning. Despite challenges such as scalability and purity control, recent advancements in automation and predictive modelling are enhancing the efficiency of combinatorial synthesis, opening new avenues for the development of metal-based catalysts, therapeutic agents, and functional materials.

Molecular modeling of cellulose tosylate immobilized α-amylases: An in silico case study through MD simulation and refinement

The use of enzymes as catalysts in industrial processes has been studied, and they offer more ecologically friendly options for chemical reactions. In the current work, we investigated the potential of molecular modeling to solve the ordinarily difficult problem of identifying the amino acids involved in the covalent mode of immobilization by in silico investigations. The immobilized α-Amylase on Cellulose tosylate (henceforth referred to as Celltos) shows extra peaks of OH and NH2, CN, SO, C-O-C, and CS. Celltos exhibits distinct ether, imine, and CS peaks, indicating the potential contribution of α-Amylase's hydroxyl, amino, and thiol groups towards immobilization with cellulose's tosylate group. The native amylase was processed for Molecular Dynamics simulation. The simulated amylase was found to be the root mean squarely deviated to 1.16 Å. Autodock Vina, GOLD, SwissDock, and iGemdock generate output averages of 6.164, 6.549, 9.313 & 137.811 and 5.903, 7.656, 9.752 & 132.218 for an unrefined and refined dataset, respectively. The catalytic site intactness values for unrefined and refined SAT9, SAT13, and LAT21 were 83.3 %, 100 %, 100 %, and 8.33 %, 0 %, and 0 %, respectively. Our findings were additionally confirmed by bond distance similarity computations.

Construction of a reaction-based fluorescent sensor for tandem detection of Cu2+ and glutathione in wine

The purpose of this study was to develop a novel reaction-based fluorescent sensor for the detection of Cu2+ and glutathione in real wine samples. The sensor, tris-(2-pyridyl)-methylamine rhodol derivative, was synthesized and validated for the tandem and selective detection of both Cu2+ and glutathione. The sensor exhibited a strong linear correlation between fluorescence intensity and Cu2+ concentration ranging from 100 to 900 nM, while the in situ generated Cu2+ ensemble selectively detected glutathione with a robust linear response from 3 to 30 μM. The detection limits for Cu2+ and glutathione were as low as 28 nM and 0.60 μM, respectively. Additionally, the sensor enabled quantitative detection of Cu2+ and glutathione in real wine samples. This work provides the first reaction-based fluorescence sensor with an "on-off-on" fluorescence response for the tandem detection of Cu2+ and glutathione in wine, offering potential applications in food and beverage quality control.

Analyzing Atomic Interactions in Molecules as Learned by Neural Networks

While machine learning (ML) models have been able to achieve unprecedented accuracies across various prediction tasks in quantum chemistry, it is now apparent that accuracy on a test set alone is not a guarantee for robust chemical modeling such as stable molecular dynamics (MD). To go beyond accuracy, we use explainable artificial intelligence (XAI) techniques to develop a general analysis framework for atomic interactions and apply it to the SchNet and PaiNN neural network models. We compare these interactions with a set of fundamental chemical principles to understand how well the models have learned the underlying physicochemical concepts from the data. We focus on the strength of the interactions for different atomic species, how predictions for intensive and extensive quantum molecular properties are made, and analyze the decay and many-body nature of the interactions with interatomic distance. Models that deviate too far from known physical principles produce unstable MD trajectories, even when they have very high energy and force prediction accuracy. We also suggest further improvements to the ML architectures to better account for the polynomial decay of atomic interactions.

Theoretical and machine learning models for reaction-barrier predictions: acrylate and methacrylate radical reactions

We propose density functional theory (DFT)- and random forest (RF)-based theoretical and machine learning (ML) models, respectively, for predicting reaction barriers (ΔETS) using acrylate and methacrylate radical reactions as representatives. DFT is used to determine 100 transition state (TS) structures of both radicals, after which the obtained data are used to determine theoretical relationships (explained with Bell-Evans-Polanyi or Brønsted-Evans-Polanyi (BEP) and Marcus-like models) between ΔETS and stabilization energy of the product. Next, we construct several theoretical regression models for predicting ΔETS of the representative reactions based on our theoretical analyses, presenting an RF-based ML model that eases ΔETS predictions by circumventing time-consuming DFT calculations. These theoretical and RF-based ML approaches will accelerate the advancement of material development.

Enhanced sampling of robust molecular datasets with uncertainty-based collective variables

Generating a dataset that is representative of the accessible configuration space of a molecular system is crucial for the robustness of machine-learned interatomic potentials. However, the complexity of molecular systems, characterized by intricate potential energy surfaces, with numerous local minima and energy barriers, presents a significant challenge. Traditional methods of data generation, such as random sampling or exhaustive exploration, are either intractable or may not capture rare, but highly informative configurations. In this study, we propose a method that leverages uncertainty as the collective variable (CV) to guide the acquisition of chemically relevant data points, focusing on regions of configuration space where ML model predictions are most uncertain. This approach employs a Gaussian Mixture Model-based uncertainty metric from a single model as the CV for biased molecular dynamics simulations. The effectiveness of our approach in overcoming energy barriers and exploring unseen energy minima, thereby enhancing the dataset in an active learning framework, is demonstrated on alanine dipeptide and bulk silica.

Functionalization of Emulsion Interfaces: Surface Chemistry Made Liquid

Disperse systems, and emulsions in particular, are currently massively used in fields as varied as food industry, cosmetics, health care and environmentally-friendly materials. To meet increasingly precise needs or targeted applications, these systems need to be endowed with new functionalities at their interfaces, in addition to their composition and structural properties. However, due to the fragility of drops and the low reactivity of their surface, conventional solid surface chemistry cannot be used for such a purpose. Several specific emulsion interface functionalization techniques have thus been developed for targeted systems and applications, but a general framework has yet to be drawn. In this review, we attempt to present these methods in a unified way through the prism of what we may call "liquid surface chemistry". We propose to categorize existing methods into drop-coating strategies, including layer-by-layer techniques and polymer coating, with a particular focus on polydopamine, and emulsifier-carrier approaches involving particles and/or amphiphilic molecules. They are discussed in a transversal way, highlighting the underlying physico-chemical principles and providing a comparative analysis of their advantages, current limitations and potential for improvement. We also propose future directions and opportunities, involving for instance DNA-based programmability or artificial intelligence, which could make liquid surface chemistry more versatile and controlled.

The design space of E(3)-equivariant atom-centred interatomic potentials

Molecular dynamics simulation is an important tool in computational materials science and chemistry, and in the past decade it has been revolutionized by machine learning. This rapid progress in machine learning interatomic potentials has produced a number of new architectures in just the past few years. Particularly notable among these are the atomic cluster expansion, which unified many of the earlier ideas around atom-density-based descriptors, and Neural Equivariant Interatomic Potentials (NequIP), a message-passing neural network with equivariant features that exhibited state-of-the-art accuracy at the time. Here we construct a mathematical framework that unifies these models: atomic cluster expansion is extended and recast as one layer of a multi-layer architecture, while the linearized version of NequIP is understood as a particular sparsification of a much larger polynomial model. Our framework also provides a practical tool for systematically probing different choices in this unified design space. An ablation study of NequIP, via a set of experiments looking at in- and out-of-domain accuracy and smooth extrapolation very far from the training data, sheds some light on which design choices are critical to achieving high accuracy. A much-simplified version of NequIP, which we call BOTnet (for body-ordered tensor network), has an interpretable architecture and maintains its accuracy on benchmark datasets.

ArcaNN: automated enhanced sampling generation of training sets for chemically reactive machine learning interatomic potentials

The emergence of artificial intelligence is profoundly impacting computational chemistry, particularly through machine-learning interatomic potentials (MLIPs). Unlike traditional potential energy surface representations, MLIPs overcome the conventional computational scaling limitations by offering an effective combination of accuracy and efficiency for calculating atomic energies and forces to be used in molecular simulations. These MLIPs have significantly enhanced molecular simulations across various applications, including large-scale simulations of materials, interfaces, chemical reactions, and beyond. Despite these advances, the construction of training datasets-a critical component for the accuracy of MLIPs-has not received proportional attention, especially in the context of chemical reactivity, which depends on rare barrier-crossing events that are not easily included in the datasets. Here we address this gap by introducing ArcaNN, a comprehensive framework designed for generating training datasets for reactive MLIPs. ArcaNN employs a concurrent learning approach combined with advanced sampling techniques to ensure an accurate representation of high-energy geometries. The framework integrates automated processes for iterative training, exploration, new configuration selection, and energy and force labeling, all while ensuring reproducibility and documentation. We demonstrate ArcaNN's capabilities through two paradigm reactions: a nucleophilic substitution and a Diels-Alder reaction. These examples showcase its effectiveness, the uniformly low error of the resulting MLIP everywhere along the chemical reaction coordinate, and its potential for broad applications in reactive molecular dynamics. Finally, we provide guidelines for assessing the quality of MLIPs in reactive systems.

Deep learning-driven prediction of chemical addition patterns for carboncones and fullerenes

Carboncones and fullerenes are exemplary π-conjugated carbon nanomaterials with unsaturated, positively curved surfaces, enabling the attachment of atoms or functional groups to enhance their physicochemical properties. However, predicting and understanding the addition patterns in functionalized carboncones and fullerenes are extremely challenging due to the formidable complexity of the regioselectivity exhibited in the adducts. Existing predictive models fall short in systems where the carbon molecular framework undergoes severe distortion upon high degrees of addition. Here, we propose an incremental deep learning approach to predict regioselectivity in the hydrogenation of carboncones and chlorination of fullerenes. Utilizing exclusively graph-based features, our deep neural network (DNN) models rely solely on atomic connectivity, without requiring 3D molecular coordinates as input or their iterative optimization. This advantage inherently avoids the risk of obtaining chemically unreasonable optimized structures, enabling the handling of highly distorted adducts. The DNN models allow us to study regioselectivity in hydrogenated carboncones of C70H20 and C62H16, accommodating up to at least 40 and 30 additional H atoms, respectively. Our approach also correctly predicts experimental addition patterns in C50Cl10 and C76Cln (n = 18, 24, and 28), whereas in the latter cases all other known methods have been proven unsuccessful. Compared to our previously developed topology-based models, the DNN's superior predictive power and generalization ability make it a promising tool for investigating complex addition patterns in similar chemical systems.

Are we fitting data or noise? Analysing the predictive power of commonly used datasets in drug-, materials-, and molecular-discovery

Data-driven techniques for establishing quantitative structure property relations are a pillar of modern materials and molecular discovery. Fuelled by the recent progress in deep learning methodology and the abundance of new algorithms, it is tempting to chase benchmarks and incrementally build ever more capable machine learning (ML) models. While model evaluation has made significant progress, the intrinsic limitations arising from the underlying experimental data are often overlooked. In the chemical sciences data collection is costly, thus datasets are small and experimental errors can be significant. These limitations of such datasets affect their predictive power, a fact that is rarely considered in a quantitative way. In this study, we analyse commonly used ML datasets for regression and classification from drug discovery, molecular discovery, and materials discovery. We derived maximum and realistic performance bounds for nine such datasets by introducing noise based on estimated or actual experimental errors. We then compared the estimated performance bounds to the reported performance of leading ML models in the literature. Out of the nine datasets and corresponding ML models considered, four were identified to have reached or surpassed dataset performance limitations and thus, they may potentially be fitting noise. More generally, we systematically examine how data range, the magnitude of experimental error, and the number of data points influence dataset performance bounds. Alongside this paper, we release the Python package NoiseEstimator and provide a web-based application for computing realistic performance bounds. This study and the resulting tools will help practitioners in the field understand the limitations of datasets and set realistic expectations for ML model performance. This work stands as a reference point, offering analysis and tools to guide development of future ML models in the chemical sciences.

Spiers Memorial Lecture: How to do impactful research in artificial intelligence for chemistry and materials science

Machine learning has been pervasively touching many fields of science. Chemistry and materials science are no exception. While machine learning has been making a great impact, it is still not reaching its full potential or maturity. In this perspective, we first outline current applications across a diversity of problems in chemistry. Then, we discuss how machine learning researchers view and approach problems in the field. Finally, we provide our considerations for maximizing impact when researching machine learning for chemistry.

Machine Learning Mapping Approach for Computing Spin Relaxation Dynamics

In this work, a machine learning mapping approach for predicting the properties of atomistic systems is reported. Within this approach, the atomic orbital overlap, density, or Kohn-Sham (KS) Fock matrix elements obtained at a low level of theory such as extended tight-binding have been used as input features to predict the electric field gradient (EFG) tensors at a higher level of theory such as those obtained with hybrid functionals. It is shown that the machine-learning-predicted EFG tensors can be used to compute spin relaxation rates of several ions in aqueous solutions. From only a fraction of data used in direct calculation, one can predict the quadrupolar isotropic spin relaxation rates with good accuracy, achieving relative errors between about 2-8% for different ions.

Concluding remarks: Faraday Discussion on NMR crystallography

This Faraday Discussion explored the field of NMR crystallography, and considered recent developments in experimental and theoretical approaches, new advances in machine learning and in the generation and handling of large amounts of data. Applications to a wide range of disordered, amorphous and dynamic systems demonstrated the range and quality of information available from this approach and the challenges that are faced in exploiting automation and developing best practice. In these closing remarks I will reflect on the discussions on the current state of the art, questions about what we want from these studies, how accurate we need results to be, how we best generate models for complex materials and what machine learning approaches can offer. These remarks close with thoughts about the future direction of the field, who will be carrying out this type of research, how they might be doing it and what their focus will be, along with likely possible challenges and opportunities.

Automatic feature selection and weighting in molecular systems using Differentiable Information Imbalance

Feature selection is essential in the analysis of molecular systems and many other fields, but several uncertainties remain: What is the optimal number of features for a simplified, interpretable model that retains essential information? How should features with different units be aligned, and how should their relative importance be weighted? Here, we introduce the Differentiable Information Imbalance (DII), an automated method to rank information content between sets of features. Using distances in a ground truth feature space, DII identifies a low-dimensional subset of features that best preserves these relationships. Each feature is scaled by a weight, which is optimized by minimizing the DII through gradient descent. This allows simultaneously performing unit alignment and relative importance scaling, while preserving interpretability. DII can also produce sparse solutions and determine the optimal size of the reduced feature space. We demonstrate the usefulness of this approach on two benchmark molecular problems: (1) identifying collective variables that describe conformations of a biomolecule, and (2) selecting features for training a machine-learning force field. These results show the potential of DII in addressing feature selection challenges and optimizing dimensionality in various applications. The method is available in the Python library DADApy.

Complete and Efficient Covariants for Three-Dimensional Point Configurations with Application to Learning Molecular Quantum Properties

When physical properties of molecules are being modeled with machine learning, it is desirable to incorporate SO(3)-covariance. While such models based on low body order features are not complete, we formulate and prove general completeness properties for higher order methods and show that 6k - 5 of these features are enough for up to k atoms. We also find that the Clebsch-Gordan operations commonly used in these methods can be replaced by matrix multiplications without sacrificing completeness, lowering the scaling from O(l6) to O(l3) in the degree of the features. We apply this to quantum chemistry, but the proposed methods are generally applicable for problems involving three-dimensional point configurations.

Data Generation for Machine Learning Interatomic Potentials and Beyond

The field of data-driven chemistry is undergoing an evolution, driven by innovations in machine learning models for predicting molecular properties and behavior. Recent strides in ML-based interatomic potentials have paved the way for accurate modeling of diverse chemical and structural properties at the atomic level. The key determinant defining MLIP reliability remains the quality of the training data. A paramount challenge lies in constructing training sets that capture specific domains in the vast chemical and structural space. This Review navigates the intricate landscape of essential components and integrity of training data that ensure the extensibility and transferability of the resulting models. We delve into the details of active learning, discussing its various facets and implementations. We outline different types of uncertainty quantification applied to atomistic data acquisition and the correlations between estimated uncertainty and true error. The role of atomistic data samplers in generating diverse and informative structures is highlighted. Furthermore, we discuss data acquisition via modified and surrogate potential energy surfaces as an innovative approach to diversify training data. The Review also provides a list of publicly available data sets that cover essential domains of chemical space.

Ligand Many-Body Expansion as a General Approach for Accelerating Transition Metal Complex Discovery

Methods that accelerate the evaluation of molecular properties are essential for chemical discovery. While some degree of ligand additivity has been established for transition metal complexes, it is underutilized in asymmetric complexes, such as the square pyramidal coordination geometries highly relevant to catalysis. To develop predictive methods beyond simple additivity, we apply a many-body expansion to octahedral and square pyramidal complexes and introduce a correction based on adjacent ligands (i.e., the cis interaction model). We first test the cis interaction model on adiabatic spin-splitting energies of octahedral Fe(II) complexes, predicting DFT-calculated values of unseen binary complexes to within an average error of 1.4 kcal/mol. Uncertainty analysis reveals the optimal basis, comprising the homoleptic and mer symmetric complexes. We next show that the cis model (i.e., the cis interaction model solved for the optimal basis) infers both DFT- and CCSD(T)-calculated model catalytic reaction energies to within 1 kcal/mol on average. The cis model predicts low-symmetry complexes with reaction energies outside the range of binary complex reaction energies. We observe that trans interactions are unnecessary for most monodentate systems but can be important for some combinations of ligands, such as complexes containing a mixture of bidentate and monodentate ligands. Finally, we demonstrate that the cis model may be combined with Δ-learning to predict CCSD(T) reaction energies from exhaustively calculated DFT reaction energies and the same fraction of CCSD(T) reaction energies needed for the cis model, achieving around 30% of the error from using the CCSD(T) reaction energies in the cis model alone.

Did organs precede organisms in the origin of life?

Evolutionary processes acting on populations of organized molecules preceded the origin of living organisms. These prebiotic entities were independently and repeatedly produced [i.e. (re)-produced] by the assembly of their components, following an iterative process giving rise to nearly but not fully identical replicas, allowing for a prebiotic form of Darwinian evolution. Natural selection favored the more persistent assemblies, some possibly modifying their own internal structure, or even their environment, thereby acquiring function. We refer to these assemblies as proto-organs. In association with other assemblies (e.g. in a coacervate or encapsulated within a vesicle), such proto-organs could evolve and acquire a role within the collective when their coexistence favored the selection of the ensemble. Along millions of years, an extraordinarily small number of successful combinations of those proto-organs co-occurring in spatially individualizing compartments might have co-evolved forming a proto-metabolic and proto-genetic informative network, eventually leading to the selfreplication of a very few. Thus, interactions between encapsulated proto-organs would have had a much higher probability of evolving into proto-organisms than interactions among simpler molecules. Multimolecular forms evolve functions; thus, functional organs would have preceded organisms.

The Role of Transition Metal Versus Coordination Mode in Single-Atom Catalyst for Electrocatalytic Sulfur Reduction Reaction

Electrocatalytic sulfur reduction reaction (SRR) is emerging as an effective strategy to combat the polysulfide shuttling effect, which remains a critical factor impeding the practical application of the Li-S battery. Single-atom catalyst (SAC), one of the most studied catalytic materials, has shown considerable potential in addressing the polysulfide shuttling effect in a Li-S battery. However, the role played by transition metal vs coordination mode in electrocatalytic SRR is trial-and-error, and the general understanding that guides the synthesis of the specific SAC with desired property remains elusive. Herein, we use first-principles calculations and machine learning to screen a comprehensive data set of graphene-based SACs with different transition metals, heteroatom doping, and coordination modes. The results reveal that the type of transition metal plays the decisive role in SAC for electrocatalytic SRR, rather than the coordination mode. Specifically, the 3d transition metals exhibit admirable electrocatalytic SRR activity for all of the coordination modes. Compared with the reported N3C1 and N4 coordinated graphene-based SACs covering 3d, 4d, and 5d transition metals, the proposed para-MnO2C2 and para-FeN2C2 possess significant advantages on the electrocatalytic SRR, including a considerably low overpotential down to 1 mV and reduced Li2S decomposition energy barrier, both suggesting an accelerated conversion process among the polysulfides. This study may clarify some understanding of the role played by transition metal vs coordination mode for SAC materials with specific structure and desired catalytic properties toward electrocatalytic SRR and beyond.

Fundamentals of the recycling of spent lithium-ion batteries

This review discusses the critical role of fundamentals of battery recycling in addressing the challenges posed by the increasing number of spent lithium-ion batteries (LIBs) due to the widespread use of electric vehicles and portable electronics, by providing the theoretical basis and technical support for recycling spent LIBs, including battery classification, ultrasonic flaw detection, pretreatment (e.g., discharging, mechanical crushing, and physical separation), electrolyte recovery, direct regeneration, and theoretical calculations and simulations. Physical chemistry principles are essential for achieving effective separation of different components through methods like screening, magnetic separation, and flotation. Electrolyte recovery involves separation and purification of electrolytes through advanced physical and chemical techniques. Direct regeneration technology restores the structure of electrode materials at the microscopic scale, requiring precise control of the physical state and crystal structure of the material. Physical processes such as phase changes, solubility, and diffusion are fundamental to techniques like solid-state sintering, eutectic-salt treatment, and hydrothermal methods. Theoretical calculations and simulations help predict the behaviour of materials during recycling, guiding process optimization. This review provides insights into understanding and improving the recycling process, emphasizing the central role of physical chemistry principles in addressing environmental and energy issues. It is valuable for promoting innovation in spent LIB recycling processes and is expected to stimulate interest among researchers and manufacturers.

Challenges with Literature-Derived Data in Machine Learning for Yield Prediction: A Case Study on Pd-Catalyzed Carbonylation Reactions

The application of machine learning (ML) to predict reaction yields has shown remarkable accuracy when based on high-throughput computational and experimental data. However, the accuracy significantly diminishes when leveraging literature-derived data, highlighting a gap in the predictive capability of the current ML models. This study, focusing on Pd-catalyzed carbonylation reactions, reveals that even with a data set of 2512 reactions, the best-performing model reaches only an R2 of 0.51. Further investigations show that the models' effectiveness is predominantly confined to predictions within narrow subsets of data, closely related and from the same literature sources, rather than across the broader, heterogeneous data sets available in the literature. The reliance on data similarity, coupled with small sample sizes from the same sources, makes the model highly sensitive to inherent fluctuations typical of small data sets, adversely impacting stability, accuracy, and generalizability. The findings underscore the inherent limitations of current ML techniques in leveraging literature-derived data for predicting chemical reaction yields, highlighting the need for more sophisticated approaches to handle the complexity and diversity of chemical data.

Can Deep Learning Search for Exceptional Chiroptical Properties? The Halogenated [6]Helicene Case

The relationship between chemical structure and chiroptical properties is not always clearly understood. Nowadays, efforts to develop new systems with enhanced optical properties follow the trial-error method. A large number of data would allow us to obtain more robust conclusions and guide research toward molecules with practical applications. In this sense, in this work we predict the chiroptical properties of millions of halogenated [6]helicenes in terms of the rotatory strength (R). We have used DFT calculations to randomly create derivatives including from 1 to 16 halogen atoms, that were then used as a data set to train different deep neural network models. These models allow us to i) predict the Rmax for any halogenated [6]helicene with a very low computational cost, and ii) to understand the physical reasons that favour some substitutions over others. Finally, we synthesized derivatives with higher predicted Rmax obtaining excellent correlation among the values obtained experimentally and the predicted ones.

Screening of BindingDB database ligands against EGFR, HER2, Estrogen, Progesterone and NF-κB receptors based on machine learning and molecular docking

Breast cancer, the second most prevalent cancer among women worldwide, necessitates the exploration of novel therapeutic approaches. To target the four subgroups of breast cancer "hormone receptor-positive and HER2-negative, hormone receptor-positive and HER2-positive, hormone receptor-negative and HER2-positive, and hormone receptor-negative and HER2-negative" it is crucial to inhibit specific targets such as EGFR, HER2, ER, NF-κB, and PR. In this study, we evaluated various methods for binary and multiclass classification. Among them, the GA-SVM-SVM:GA-SVM-SVM model was selected with an accuracy of 0.74, an F1-score of 0.73, and an AUC of 0.92 for virtual screening of ligands from the BindingDB database. This model successfully identified 4454, 803, 438, and 378 ligands with over 90% precision in both active/inactive and target prediction for the classes of EGFR+HER2, ER, NF-κB, and PR, respectively, from the BindingDB database. Based on to the selected ligands, we created a dendrogram that categorizes different ligands based on their targets. This dendrogram aims to facilitate the exploration of chemical space for various therapeutic targets. Ligands that surpassed a 90% threshold in the product of activity probability and correct target selection probability were chosen for further investigation using molecular docking. The binding energy range for these ligands against their respective targets was calculated to be between -15 and -5 kcal/mol. Finally, based on general and common rules in medicinal chemistry, we selected 2, 3, 3, and 8 new ligands with high priority for further studies in the EGFR+HER2, ER, NF-κB, and PR classes, respectively.

Exploring Cationic Substitutions in the Solid Electrolyte NaAlCl4 with Density Functional Theory

NaAlCl4 is an established solid electrolyte in high-temperature Na-based battery systems, but its ionic conductivity is not sufficiently high for room-temperature applications. We employ density functional theory and thermodynamic corrections to evaluate the efficacy of various elements for substitution, utilizing on-the-fly machine-learned potentials to accelerate the required phonon calculations by 1 order of magnitude at a minor error of -0.7 ± 1.0 meV/atom. All investigated isovalent substitutions are favorable within 4 meV/atom, with potassium and silver as substitutes for sodium and gallium as a substitute for aluminum. The most promising aliovalent substitution was identified for Zn on the tieline between NaAlCl4 and Na2ZnCl4. The structure of latter, with aluminum ions replacing zinc, yields a structure with separate layers for the differently charged cations and vacancies for potential Na conduction. Our investigation may pave the way for more reliable discovery of new Na conductors by inclusion of thermodynamic properties.

Surface-Enhanced Raman Scattering-Based Multimodal Techniques: Advances and Perspectives

Surface-enhanced Raman scattering (SERS) spectroscopy is a versatile molecular fingerprinting technique with rapid signal readout, high aqueous compatibility, and portability. To translate SERS for real-world applications, it is pertinent to overcome inherent challenges, including high sample variability and heterogeneity, matrix effects, and nonlinear SERS signal responses of different analytes in complex (bio)chemical matrices with numerous interfering species. In this perspective, we highlight emerging SERS-based multimodal techniques to address the key roadblocks to improving the sensitivity, specificity, and reliability of (bio)chemical detection, bioimaging, theragnosis, and theragnostic. SERS-based multimodal techniques can be broadly categorized into two categories: (1) complementary methods or systems that work together to achieve a common goal where each method compensates for the weaknesses of the other to culminate in a single enhanced outcome or (2) orthogonal techniques that are independent and provide separate but corroborating results simultaneously without interfering with each other. These multimodal techniques maximize information gained from a single experiment to achieve enhanced qualitative or quantitative analysis and broaden the range of detectable analytes from small molecules to tissues. Finally, we discuss emerging directions in multimodal platform design, instrument integration, and data analytics that aim to push the analytical limits of holistic detection.

A Multipole-Based Reactive Force Field for Hydrocarbons

The computational complexity of quantum chemistry methods has prompted the development of reactive force fields, facilitating practical applications of molecular dynamics simulations for large-scale reactive systems. Current reactive force fields typically employ intricate corrections based on prior chemical knowledge, which severely impedes their further advancement. This study presents a new atomic multipole-based reactive model with bond free (OPERATOR). The force field is constructed on a simple, physically motivated model within the AMOEBA framework that closely resembles the physical representation of the chemical reaction processes. In the force field, the atomic multipoles are generated dynamically according to the atomic environments, aiming to effectively capture significant changes in the electrostatic environments during chemical reactions. Subsequently, atomic multipole-based charge penetration, polarization, and charge transfer effects are incorporated into the force field to describe the complex electrostatic interactions in the system. The force field also includes van der Waals interactions and three-body potentials. In addition, to extend these nonreactive interactions to chemical reactions, the atom distribution multipole moments are used to characterize different chemical environments. The force field has been optimized using the dataset of potential energy surfaces (PESs) of hydrocarbons derived from DFT results of millions of conformations with six degrees of freedom (DOFs). The results demonstrate that the new force field effectively replicates both the monopoles and the energies. In comparison to ReaxFF, the new force field exhibits comparable or superior performance. Furthermore, molecular dynamics simulations of n-heptane decomposition effectively reproduce the primary products and reactions observed in the experiments. Given the simplicity and physically motivated nature of the model, it is expected that the new force field will be utilized in future studies to investigate chemical reaction mechanisms involving more elements.

Data-Based Prediction of Redox Potentials via Introducing Chemical Features into the Transformer Architecture

Rapid and accurate prediction of basic physicochemical parameters of molecules will greatly accelerate the target-orientated design of novel reactions and materials but has been long challenging. Herein, a chemical language model-based deep learning method, TransChem, has been developed for the prediction of redox potentials of organic molecules. Embedding an effective molecular characterization (combining spatial and electronic features), a nonlinear molecular messaging approach (Mol-Attention), and a perturbation learning method, TransChem, shows high accuracy in predicting the redox potential of organic radicals comprising over 100,000 data (R2 > 0.97, MAE <0.09 V) and is generalized to the smaller 2,1,3-benzothiadiazole data set (<3000 data points) and electron affinity data set (660 data) with low MAE of 0.07 V and 0.18 eV, respectively. In this context, a self-developed data set, i.e., the oxidation potential (OP) of a full-space disubstituted phenol data set (OPP-data set, total set: 74,529), has been predicted by TransChem with a high-throughput, and active learning strategy. The rapid and reliable prediction of OP could hopefully accelerate the screening of plausible reagents in highly selective cross-coupling of phenol derivatives. This study presents an important attempt to guide language modeling with chemical knowledge, while TransChem demonstrates state-of-the-art (SOTA) predictive performance on redox potential prediction benchmark data sets for its better understanding of molecular design and conformational relationships.

Progress in Multiscale Modeling of Silk Materials

As a result of their hierarchical structure and biological processing, silk fibers rank among nature's most remarkable materials. The biocompatibility of silk-based materials and the exceptional mechanical properties of certain fibers has inspired the use of silk in numerous technical and medical applications. In recent years, computational modeling has clarified the relationship between the molecular architecture and emergent properties of silk fibers and has demonstrated predictive power in studies on novel biomaterials. Here, we review advances in modeling the structure and properties of natural and synthetic silk-based materials, from early structural studies of silkworm cocoon fibers to cutting-edge atomistic simulations of spider silk nanofibrils and the recent use of machine learning models. We explore applications of modeling across length scales: from quantum mechanical studies on model peptides, to atomistic and coarse-grained molecular dynamics simulations of silk proteins, to finite element analysis of spider webs. As computational power and algorithmic efficiency continue to advance, we expect multiscale modeling to become an indispensable tool for understanding nature's most impressive fibers and developing bioinspired functional materials.

Prediction of the Infrared Absorbance Intensities and Frequencies of Hydrocarbons: A Message Passing Neural Network Approach

Accurately and efficiently predicting the infrared (IR) spectra of a molecule can provide insights into the structure-properties relationships of molecular species, which has led to a proliferation of machine learning tools designed for this purpose. However, earlier studies have focused primarily on obtaining normalized IR spectra, which limits their potential for a comprehensive analysis of molecular behavior in the IR range. For instance, to fully understand and predict the optical properties, such as the transparency characteristics, it is necessary to predict the molar absorptivity IR spectra instead. Here, we propose a graph-based communicative message passing neural network algorithm that can predict both the peak positions and absolute intensities corresponding to density functional theory calculated molar absorptivities in the IR domain. By modifying existing spectral loss functions, we show that our method is able to predict with DFT-accuracy level the IR molar absorptivities of a series of hydrocarbons containing up to ten carbon atoms and apply the model to a set of larger molecules. We also compare the predicted spectra with those generated by the direct message passing neural network. The results suggest that both algorithms demonstrate similar predictive capabilities for hydrocarbons, indicating that either model could be effectively used in future research on spectral prediction for such systems.

Exploring the Crystal Structure and Electronic Properties of γ-Al2O3: Machine Learning Drives Future Material Innovations

For decades, researchers have struggled to determine the precise crystal structure of γ-Al2O3 due to its atomic-level disorder and the challenges associated with obtaining high-purity, high-crystallinity γ-Al2O3 in laboratory settings. This study investigates the crystal structure and electronic properties of γ-Al2O3 coatings under the influence of an external electric field, integrating machine learning with density functional theory (DFT). A potential 160-atom supercell structure was identified from over 600,000 γ-Al2O3 configurations and confirmed through high-resolution transmission electron microscopy and selected area electron diffraction. The findings indicate that γ-Al2O3 deviates from the conventional spinel structure, suggesting that octahedral vacancies can reduce the system's energy. Under an external electric field, the material's band structure and density of states (DOS) undergo significant changes: the bandgap narrows from 3.996 to 0 eV, resulting in metallic behavior, while the projected density of states (PDOS) exhibits peak broadening and splitting of oxygen atom PDOS below the Fermi level. These alterations elucidate the variations in the electrical conductivity of alumina coatings under an electric field. These findings clarify the mechanisms of γ-Al2O3's electronic property modulation and offer insights into its covalent and ionic mixed bonding as a wide-bandgap semiconductor. This discovery is essential for understanding dielectric breakdown in insulating materials.

Predicting DNA Reactions with a Quantum Chemistry-Based Deep Learning Model

In this study, a deep learning model based on quantum chemistry is introduced to enhance the accuracy and efficiency of predicting DNA reaction parameters. By integrating quantum chemical calculations with self-designed descriptor matrices, the model offers a comprehensive description of energy variations and considers a broad range of relevant factors. To overcome the challenge of limited labeled data, an active learning method is employed. The results demonstrate that this model outperforms existing methods in predicting DNA hybridization free energies and strand displacement rate constants, thus advancing the understanding of DNA molecular interactions, and aiding in the precise design and optimization of DNA-based systems.

Predicting redox potentials by graph-based machine learning methods

The evaluation of oxidation and reduction potentials is a pivotal task in various chemical fields. However, their accurate prediction by theoretical computations, which is a complementary task and sometimes the only alternative to experimental measurement, may be often resource-intensive and time-consuming. This paper addresses this challenge through the application of machine learning techniques, with a particular focus on graph-based methods (such as graph edit distances, graph kernels, and graph neural networks) that are reviewed to enlighten their deep links with theoretical chemistry. To this aim, we establish the ORedOx159 database, a comprehensive, homogeneous (with reference values stemming from density functional theory calculations), and reliable resource containing 318 one-electron reduction and oxidation reactions and featuring 159 large organic compounds. Subsequently, we provide an instructive overview of the good practice in machine learning and of commonly utilized machine learning models. We then assess their predictive performances on the ORedOx159 dataset through extensive analyses. Our simulations using descriptors that are computed in an almost instantaneous way result in a notable improvement in prediction accuracy, with mean absolute error (MAE) values equal to 5.6 kcal mol- 1 for reduction and 7.2 kcal mol- 1 for oxidation potentials, which paves a way toward efficient in silico design of new electrochemical systems.

Electronic structure simulations in the cloud computing environment

The transformative impact of modern computational paradigms and technologies, such as high-performance computing (HPC), quantum computing, and cloud computing, has opened up profound new opportunities for scientific simulations. Scalable computational chemistry is one beneficiary of this technological progress. The main focus of this paper is on the performance of various quantum chemical formulations, ranging from low-order methods to high-accuracy approaches, implemented in different computational chemistry packages and libraries, such as NWChem, NWChemEx, Scalable Predictive Methods for Excitations and Correlated Phenomena, ExaChem, and Fermi-Löwdin orbital self-interaction correction on Azure Quantum Elements, Microsoft's cloud services platform for scientific discovery. We pay particular attention to the intricate workflows for performing complex chemistry simulations, associated data curation, and mechanisms for accuracy assessment, which is demonstrated with the Arrows automated workflow for high throughput simulations. Finally, we provide a perspective on the role of cloud computing in supporting the mission of leadership computational facilities.

Implementation of frozen density embedding in CP2K and OpenMolcas: CASSCF wavefunctions embedded in a Gaussian and plane wave DFT environment

Most chemical processes happen at a local scale where only a subset of molecular orbitals is directly involved and only a subset of covalent bonds may be rearranged. To model such reactions, Density Functional Theory (DFT) is often inadequate, and the use of computationally more expensive correlated wavefunction (WF) methods is required for accurate results. Mixed-resolution approaches backed by quantum embedding theory have been used extensively to approach this imbalance. Based on the frozen density embedding freeze-and-thaw algorithm, we describe an approach to embed complete active space self-consistent field simulations run in the OpenMolcas code in a DFT environment calculated in CP2K without requiring any external tools. This makes it possible to study a local, active part of a chemical system in a larger and relatively static environment with a computational cost balanced between the accuracy of a WF method and the efficiency of DFT, which we test on environment-subsystem pairs. Finally, we apply the implementation to an oxygen molecule leaving an aluminum (111) surface and a ruthenium(IV) oxide (110) surface.

Accurate modeling of the potential energy surface of atmospheric molecular clusters boosted by neural networks

The computational cost of accurate quantum chemistry (QC) calculations of large molecular systems can often be unbearably high. Machine learning offers a lower computational cost compared to QC methods while maintaining their accuracy. In this study, we employ the polarizable atom interaction neural network (PaiNN) architecture to train and model the potential energy surface of molecular clusters relevant to atmospheric new particle formation, such as sulfuric acid-ammonia clusters. We compare the differences between PaiNN and previous kernel ridge regression modeling for the Clusteromics I-V data sets. We showcase three models capable of predicting electronic binding energies and interatomic forces with mean absolute errors of <0.3 kcal mol-1 and <0.2 kcal mol-1 Å-1, respectively. Furthermore, we demonstrate that the error of the modeled properties remains below the chemical accuracy of 1 kcal mol-1 even for clusters vastly larger than those in the training database (up to (H2SO4)15(NH3)15 clusters, containing 30 molecules). Consequently, we emphasize the potential applications of these models for faster and more thorough configurational sampling and for boosting molecular dynamics studies of large atmospheric molecular clusters.

Exploring 7β-amino-6-nitrocholestens as COVID-19 antivirals: in silico, synthesis, evaluation, and integration of artificial intelligence (AI) in drug design: assessing the cytotoxicity and antioxidant activity of 3β-acetoxynitrocholestane

In light of the ongoing pandemic caused by SARS-CoV-2, effective and clinically translatable treatments are desperately needed for COVID-19 and its emerging variants. In this study, some derivatives, including 7β-aminocholestene compounds, and 3β-acetoxy-6-nitrocholesta-4,6-diene were synthesized, in quantitative yields from 7β-bromo-6-nitrocholest-5-enes (1-3) with a small library of amines. The synthesized steroidal products were then thoroughly characterized using a range of physicochemical techniques, including IR, NMR, UV, MS, and elemental analysis. Next, a virtual screening based on structures using docking studies was conducted to investigate the potential of these synthesized compounds as therapeutic candidates against SARS-CoV-2. Specifically, we evaluated the compounds' binding energy of the reactants and their products with three SARS-CoV-2 functional proteins: the papain-like protease, 3C-like protease or main protease, and RNA-dependent RNA polymerase. Our results indicate that the 7β-aminocholestene derivatives (4-8) display intermediate to excellent binding energy, suggesting that they interact strongly with the receptor's active amino acids and may be promising drug candidates for inhibiting SARS-CoV-2. Although the starting steroid derivatives; 7β-bromo-6-nitrocholest-5-enes (1-3) and one steroid product; 3β-acetoxy-6-nitrocholesta-4,6-diene (9) exhibited strong binding energies with various SARS-CoV-2 receptors, they did not meet the Lipinski Rule and ADMET properties required for drug development. These compounds showed either mutagenic or reproductive/developmental toxicity when assessed using toxicity prediction software. The findings based on structure-based virtual screening, suggest that 7β-aminocholestaines (4-8) may be useful for reducing the susceptibility to SARS-CoV-2 infection. The docking pose of compound 4, which has a high score of -7.4 kcal mol-1, was subjected to AI-assisted deep learning to generate 60 AI-designed molecules for drug design. Molecular docking of these AI molecules was performed to select optimal candidates for further analysis and visualization. The cytotoxicity and antioxidant effects of 3β-acetoxy-6-nitrocholesta-4,6-diene were tested in vitro, showing marked cytotoxicity and antioxidant activity. To elucidate the molecular basis for these effects, steroidal compound 9 was subjected to molecular docking analysis to identify potential binding interactions. The stability of the top-ranked docking pose was subsequently assessed using molecular dynamics simulations.