Themed collection Machine Learning and Artificial Intelligence: A cross-journal collection

By integrating machine learning with automation and robots, accelerated discovery of photocatalysts in the future could be envisioned.

Many of the tasks in the molecular design pipeline can be modelled as inverse problems. This highlight focuses on recent developments in modern machine learning methods which can be used to tackle those inverse problems.

Artificial Neural Networks (NN) are already heavily involved in methods and applications for frequent tasks in the field of computational chemistry such as representation of potential energy surfaces (PES) and spectroscopic predictions.

This review summarises recent advances in the use of machine learning for predicting friction and wear in tribological systems, material discovery, lubricant design and composite formulation. Potential future applications and areas for further research are also discussed.

The machine learning atomic simulation will usher the research of zeolite, as other complex materials, into a new era featuring the easy access to zeolite functionalities predicted from theory.

This article provides a vision on how to accelerate the production of chemicals and fuels from biomass feedstocks using an integrated framework of data mining, retrosynthesis, lab automation, and process systems engineering.

Enzyme engineering is an important biotechnological process capable of generating tailored biocatalysts for applications in industrial chemical conversion and biopharma.

This review explores machine learning's role in energy chemistry, spanning organic photovoltaics, perovskites, catalysis, and batteries, highlighting its potential to accelerate eco-friendly, sustainable energy development.

This paper reviews the research progress in memristor-based neural networks and puts forward future development trends.

Data-driven machine learning is a proven technique for battery material discovery and enables the development of sustainable next-generation batteries.

In this review, we focus on the systematic construction of the data-driven electrocatalyst design framework and discuss its principles, current challenges, and opportunities.

Machine learning can accelerate the process of electrocatalyst discovery and optimization, especially when incorporated into a closed-loop approach with autonomous laboratories. This review highlights the recent progress and challenges in this field.

This review outlines several organic chemistry tasks for which predictive machine learning models have been and can be applied.

The cross-fertilisation between the listed disciplines with a long standing knowledge on the application of artificial intelligence protocols and electron microscopy for materials science can entail the next breakthroughs in the field.

Quantum variants of machine learning algorithms are discussed with emphasis on methodology, learning techniques and applications in broad and distinct domains of chemical physics.

Understand and optimize industrial processes via machine learning and chemical engineering principles.

The paradigm of advanced materials has grown exponentially over the last decade, with their new dimensions including digital design, dynamics, and functions.

Machine learning-assisted synthesis of nanoparticles.

The perspective of machine learning for modeling plasma treatment parameters in agriculture for the development of synergistic protocols for different types of seed priming.

Machine learning (ML) is becoming an effective tool for studying 2D materials.

Within the agri-food sector, e-noses, e-tongues, and e-eyes generate interest. This review delves into their principles, challenges, and data processing methods, featuring case studies that contribute to the advancement of e-sensing technologies.

Machine learning applications in microalgae biofuel production are reviewed; the current states and major trends in research as well as the challenges to overcome are identified.

In this review article, we surveyed the applications of machine learning in microfluidic design and microfluidic control.

A multi-step and multi-stage high-throughput materials search via ensemble machine learning is developed to carefully and comprehensively screen nearly 12 million all-inorganic perovskites with potential high photovoltaic performance.

By training 50 fundamental organic reactions, the learning model predicted the products and pathways of 35 test reactions. The model identified the key fragment structures of the reaction intermediates.

Rapid prediction of environmental chemistry properties is critical for the green and sustainable development of the chemical industry and drug discovery.

A graph-based machine learning model is built to predict atom dynamics from their static structure, which, in turn, unveils the predictive power of static structure in dynamical evolution of disordered phases.

Schematic of our bidirectional, ML-empowered approach incorporating plasmonic featurization for rapid (<30 s) and accurate determination of the size and size distribution of gold nanosphere (Au NSs) ensembles in real samples.

We present a hybrid neural network and genetic optimization adaptive method incorporating Bézier curves to consider the large design space of lattice structures with superior mechanical properties.

Fine-tuned GPT-3 shows robust performance for the prediction of electronic and functional properties for organic molecules, with resilience to information loss and noise.

The dominant binding mode of the QUB-00006-Int-07 main protease inhibitor during absolute binding free energy simulations.

Contrastive pretraining of chemical reactions by matching augmented reaction representations to improve machine learning performance on small reaction datasets.

An automated flow chemistry platform was designed to collect data for a lithium-halogen exchange reaction. The data was used to train a Bayesian multi-objective optimization algorithm to optimize the process parameters and build process knowledge.

Cellular autofluorescence signatures, considered to represent the physiological state of individual cells, allow us to discriminate mycoplasma infection using machine learning models.

Machine learning is increasingly popular and promising in environmental science due to its potential in solving various environmental problems, particularly with simple code-free tools.

We classify native and chemically modified RBCs with an AI based video classifier at high accuracy (>90%). We use chemicals to mimic aspects of typical RBC disorders. This enables a label-free categorization, based on cell shape and flow dynamics.

The semi-supervised machine learning approach is an integrated calibration-free modelling framework for identifying reaction systems from spectral data using minimal prior information and it is validated with experimental data obtained in a micro-reactor.

An automated and intelligent microfluidic platform (AIMP), which offers automated system control, intelligent data analysis, and user interaction was developed to provide a cost-effective and portable solution for detecting and monitoring microalgae.

New MODA descriptor, a quantum-inspired representation enhancing ML predictions of molecular properties. By using a wave-function guess, MODA captures electronic structure intricacies to excel in intermolecular property predictions.

We report a machine learning method for predicting the atom pair correlation functions of a class of glassy materials.

We used machine learning to predict the self-assembly structures of amphiphilic molecules and analyzed the physical factors affecting their morphologies.

Solubility prediction plays a crucial role in energy storage applications, such as redox flow batteries, because it directly affects the efficiency and reliability.

We show that combining a variety of public datasets and non-linear machine learning models can predict emissions from electric generating units at good accuracy without any proprietary information.

The data-driven machine learning approach has greatly improved the predictive accuracy of T_g and D_max values. The governing rules for GFA have been successfully established through feature significance analysis.

A machine-learning methodology was applied to unveil the structure–property relationships of the fabricated ternary Ni, Fe, and Co amorphous oxygen evolution catalyst, showcasing remarkable performance and stability via corrosion engineering.

In this article, a bidirectional clustering model proposed for methanol-reforming catalysts demonstrates excellent mathematical performance and is of significance for the discovery of methanol-reforming catalysts.

Polycyclic aromatic hydrocarbons (PAHs) have attracted a lot of environmental concern because of their carcinogenic and mutagenic properties, and the fact they can easily contaminate natural resources such as drinking water and river water.

Ionic liquids have many intriguing properties and widespread applications such as separations and energy storage.

We experimentally justify the advantages of jumping on the deep learning trend for image-activated budding yeast sorting and validate its applicability towards morphology-based yeast mutant screening.

Vaccination and infection rates against variants of COVID-19 in Dane County, WI were determined from low-volume human sera/plasma samples with machine-learning aided nanoplasmonic biosensor. The results agree with the official epidemiological data.

Unveiling current issues in the investigation of highly-active heterogeneous catalysts using machine learning engineering techniques was discussed in the case of oxidative coupling of methane with support vector regression and Bayesian optimization.

A machine learning-based nano-photocatalyst module for accelerating the design of Bi₂WO₆/MIL-53(Al) nanocomposites was constructed by four steps. An online web service was established to quickly predict the photocatalytic activity of Bi₂WO₆/MIL-53(Al).

Data-driven predictions of metal hydride thermodynamic properties elucidate the Pareto optimal front of high entropy alloy candidates for hydrogen storage.

We report an ML pipeline that predicts the nanoparticle–nanoparticle pair correlation function of a polymer nanocomposite.

Application of a machine learning approach to device design. Starting from database analysis followed by a dataset creation based on those insights. Data preprocessing is done to extract features for ML prediction and design new PSCs.

The effective passivation material (ITIC) for P–I–N type perovskite solar cells is selected by machine learning. In the verification experiment, the defect density of the perovskite layer is significantly decreased after treatment with ITIC.

The deployment of microneedles in biological fluid sampling and drug delivery is an emerging field in biotechnology, which contributes greatly to minimally-invasive methods in medicine.

A Machine Learning (ML) approach was proposed to optimize the manufacturing-route selection from the physical particle properties of a pharmaceutical material.

A prediction model for cloud point was built by a combination of materials informatics and chemical insight.

We report the development of machine learning model for the calculation of carbon dioxide solubilities in deep solvent solvents. This model helps to predict and accelerate the development of carbon capture solvents with ideal experimental conditions.

Utilizes machine learning to propose an absorption bandwidth and structural parameters prediction approach for the design of patterned graphene metasurface absorber, which provides a new direction for the precision design of optical devices.

A novel semi-automated, high-throughput computational workflow for ligand/catalyst discovery based on the Cambridge Structural Database is reported.

A fast machine learning based framework is introduced for the prediction of solubility parameters and selection of green solvents for small molecular donor-based organic solar cells.

We introduce a machine-learning-assisted workflow to write, read, and classify dye-loaded PLGA–PEG nanoparticles at a single-particle level.

We establish a robust and broadly applicable multistep workflow using machine learning algorithms to construct well-trained data-driven models for predicting the hydrogen evolution reaction activity of 4500 MM′XT₂-type MXenes.

Prediction of physical observables with machine learning for spintronic and molecular devices.

Machine learning motivated Cu@MoS₂ catalysts design for enhanced peroxidase-like activity.

Using ridge regression, the propagation rate coefficients for radical polymerization are correlated with basic molecular properties.

CO₂RR binary alloy catalyst design insight gained through density functional theory and machine learning with a focus on COCOH adsorption energy.

Artificial neural networks trained on 23 density functional approximations (DFAs) from multiple rungs of “Jacob's ladder” enable the prediction of where each DFA has zero curvature for chemical discovery.