Message-passing neural networks for high-throughput polymer screening

Transfer learning across different chemical domains: virtual screening of organic materials with deep learning models pretrained on small molecule and chemical reaction data

Machine learning is becoming a preferred method for the virtual screening of organic materials due to its cost-effectiveness over traditional computationally demanding techniques. However, the scarcity of labeled data for organic materials poses a significant challenge for training advanced machine learning models. This study showcases the potential of utilizing databases of drug-like small molecules and chemical reactions to pretrain the BERT model, enhancing its performance in the virtual screening of organic materials. By fine-tuning the BERT models with data from five virtual screening tasks, the version pretrained with the USPTO-SMILES dataset achieved R2 scores exceeding 0.94 for three tasks and over 0.81 for two others. This performance surpasses that of models pretrained on the small molecule or organic materials databases and outperforms three traditional machine learning models trained directly on virtual screening data. The success of the USPTO-SMILES pretrained BERT model can be attributed to the diverse array of organic building blocks in the USPTO database, offering a broader exploration of the chemical space. The study further suggests that accessing a reaction database with a wider range of reactions than the USPTO could further enhance model performance. Overall, this research validates the feasibility of applying transfer learning across different chemical domains for the efficient virtual screening of organic materials.Scientific contributionThis study verifies the feasibility of applying transfer learning to large language models in different chemical fields to help organic materials perform virtual screening. Through the comparison of transfer learning from different chemical fields to a variety of organic material molecules, the high precision virtual screening of organic materials is realized.

AI-guided few-shot inverse design of HDP-mimicking polymers against drug-resistant bacteria

Host defense peptide (HDP)-mimicking polymers are promising therapeutic alternatives to antibiotics and have large-scale untapped potential. Artificial intelligence (AI) exhibits promising performance on large-scale chemical-content design, however, existing AI methods face difficulties on scarcity data in each family of HDP-mimicking polymers (<102), much smaller than public polymer datasets (>105), and multi-constraints on properties and structures when exploring high-dimensional polymer space. Herein, we develop a universal AI-guided few-shot inverse design framework by designing multi-modal representations to enrich polymer information for predictions and creating a graph grammar distillation for chemical space restriction to improve the efficiency of multi-constrained polymer generation with reinforcement learning. Exampled with HDP-mimicking β-amino acid polymers, we successfully simulate predictions of over 105 polymers and identify 83 optimal polymers. Furthermore, we synthesize an optimal polymer DM0.8iPen0.2 and find that this polymer exhibits broad-spectrum and potent antibacterial activity against multiple clinically isolated antibiotic-resistant pathogens, validating the effectiveness of AI-guided design strategy.

Accelerating Polymer Discovery with Uncertainty-Guided PGCNN: Explainable AI for Predicting Properties and Mechanistic Insights

Deep learning holds great potential for expediting the discovery of new polymers from the vast chemical space. However, accurately predicting polymer properties for practical applications based on their monomer composition has long been a challenge. The main obstacles include insufficient data, ineffective representation encoding, and lack of explainability. To address these issues, we propose an interpretable model called the Polymer Graph Convolutional Neural Network (PGCNN) that can accurately predict various polymer properties. This model is trained using the RadonPy data set and validated using experimental data samples. By integrating evidential deep learning with the model, we can quantify the uncertainty of predictions and enable sample-efficient training through uncertainty-guided active learning. Additionally, we demonstrate that the global attention of the graph embedding can aid in discovering underlying physical principles by identifying important functional groups within polymers and associating them with specific material attributes. Lastly, we explore the high-throughput screening capability of our model by rapidly identifying thousands of promising candidates with low and high thermal conductivity from a pool of one million hypothetical polymers. In summary, our research not only advances our mechanistic understanding of polymers using explainable AI but also paves the way for data-driven trustworthy discovery of polymer materials.

Synergy of machine learning and density functional theory calculations for predicting experimental Lewis base affinity and Lewis polybase binding atoms

Investigation of Lewis acid-base interactions has been conducted by ab initio calculations and machine learning (ML) models. This study aims to resolve two critical tasks that have not been quantitatively investigated. First, ML models developed from density functional theory (DFT) calculations predict experimental BF3 affinity with Pearson correlation coefficients around 0.9 and mean absolute errors around 10 kJ mol-1. The ML models are trained by DFT-calculated BF3 affinity of more than 3000 adducts, with input features readily obtained by rdkit. Second, the ML models have the capability of predicting the relative strength of Lewis base binding atoms in Lewis polybases, which is either an extremely challenging task to conduct experimentally or a computationally expensive task for ab initio methods. The study demonstrates and solidifies the potential of combining DFT calculations and ML models to predict experimental properties, especially those that are scarce and impractical to empirically acquire.

Molecular hypergraph neural networks

Graph neural networks (GNNs) have demonstrated promising performance across various chemistry-related tasks. However, conventional graphs only model the pairwise connectivity in molecules, failing to adequately represent higher order connections, such as multi-center bonds and conjugated structures. To tackle this challenge, we introduce molecular hypergraphs and propose Molecular Hypergraph Neural Networks (MHNNs) to predict the optoelectronic properties of organic semiconductors, where hyperedges represent conjugated structures. A general algorithm is designed for irregular high-order connections, which can efficiently operate on molecular hypergraphs with hyperedges of various orders. The results show that MHNN outperforms all baseline models on most tasks of organic photovoltaic, OCELOT chromophore v1, and PCQM4Mv2 datasets. Notably, MHNN achieves this without any 3D geometric information, surpassing the baseline model that utilizes atom positions. Moreover, MHNN achieves better performance than pretrained GNNs under limited training data, underscoring its excellent data efficiency. This work provides a new strategy for more general molecular representations and property prediction tasks related to high-order connections.

Gibbs-Helmholtz Graph Neural Network for the Prediction of Activity Coefficients of Polymer Solutions at Infinite Dilution

Machine learning models have gained prominence for predicting pure-component properties, yet their application to mixture property prediction remains relatively limited. However, the significance of mixtures in our daily lives is undeniable, particularly in industries such as polymer processing. This study presents a modification of the Gibbs-Helmholtz graph neural network (GH-GNN) model for predicting weight-based activity coefficients at infinite dilution (Ωij∞) in polymer solutions. We evaluate various polymer representations ranging from monomer, repeating unit, periodic unit, and oligomer and observe that, in data-scarce scenarios of polymer-solvent mixtures, polymer representation specifics have a reduced impact compared to data-rich environments. Leveraging transfer learning, we harness richer activity coefficient data from small-size systems, enhancing model accuracy and reducing prediction variability. The modified GH-GNN model achieves remarkable prediction results in mixture interpolation and solvent extrapolation tasks having an overall mean absolute error of 0.15, showcasing the potential of graph-neural-network-based models for property prediction of polymer solutions. Comparative analysis with the established models UNIFAC-ZM and Entropic-FV suggests a promising avenue for future research on the use of data-driven models for the prediction of the thermodynamic properties of polymer solutions.

Prediction and Interpretability of Glass Transition Temperature of Homopolymers by Data-Augmented Graph Convolutional Neural Networks

Establishing the structure-property relationship by machine learning (ML) models is extremely valuable for accelerating the molecular design of polymers. However, existing ML models for the polymers are subject to scarcity issues of training data and fewer variations of graph structures of molecules. In addition, limited works have explored the interpretability of ML models to infer the latent knowledge in the field of polymer science that could inspire ML-assisted molecular design. In this contribution, we integrate graph convolutional neural networks (GCNs) with data augmentation strategy to predict the glass transition temperature Tg of polymers. It is demonstrated that the data-augmented GCN model outperforms the conventional models and achieves a higher accuracy for the prediction of Tg despite a small amount of training data. Furthermore, taking advantage of molecular graph representations, the data-augmented GCN model has the capability to infer the importance of atoms or substructures from the understanding of Tg, which generally agrees with the experimental findings in the field of polymer science. The inferred knowledge of the GCN model is used to advise on the design of functional polymers with specific Tg. The data-augmented GCN model possesses prominent superiorities in the establishment of structure-property relationship and also provides an efficient way for accelerating the rational design of polymer molecules.

Variational autoencoder-based chemical latent space for large molecular structures with 3D complexity

The structural diversity of chemical libraries, which are systematic collections of compounds that have potential to bind to biomolecules, can be represented by chemical latent space. A chemical latent space is a projection of a compound structure into a mathematical space based on several molecular features, and it can express structural diversity within a compound library in order to explore a broader chemical space and generate novel compound structures for drug candidates. In this study, we developed a deep-learning method, called NP-VAE (Natural Product-oriented Variational Autoencoder), based on variational autoencoder for managing hard-to-analyze datasets from DrugBank and large molecular structures such as natural compounds with chirality, an essential factor in the 3D complexity of compounds. NP-VAE was successful in constructing the chemical latent space from large-sized compounds that were unable to be handled in existing methods, achieving higher reconstruction accuracy, and demonstrating stable performance as a generative model across various indices. Furthermore, by exploring the acquired latent space, we succeeded in comprehensively analyzing a compound library containing natural compounds and generating novel compound structures with optimized functions.

PolyID: Artificial Intelligence for Discovering Performance-Advantaged and Sustainable Polymers

A necessary transformation for a sustainable economy is the transition from fossil-derived plastics to polymers derived from biomass and waste resources. While renewable feedstocks can enhance material performance through unique chemical moieties, probing the vast material design space by experiment alone is not practically feasible. Here, we develop a machine-learning-based tool, PolyID, to reduce the design space of renewable feedstocks to enable efficient discovery of performance-advantaged, biobased polymers. PolyID is a multioutput, graph neural network specifically designed to increase accuracy and to enable quantitative structure-property relationship (QSPR) analysis for polymers. It includes a novel domain-of-validity method that was developed and applied to demonstrate how gaps in training data can be filled to improve accuracy. The model was benchmarked with both a 20% held-out subset of the original training data and 22 experimentally synthesized polymers. A mean absolute error for the glass transition temperatures of 19.8 and 26.4 °C was achieved for the test and experimental data sets, respectively. Predictions were made on polymers composed of monomers from four databases that contain biologically accessible small molecules: MetaCyc, MINEs, KEGG, and BiGG. From 1.4 × 106 accessible biobased polymers, we identified five poly(ethylene terephthalate) (PET) analogues with predicted improvements to thermal and transport performance. Experimental validation for one of the PET analogues demonstrated a glass transition temperature between 85 and 112 °C, which is higher than PET and within the predicted range of the PolyID tool. In addition to accurate predictions, we show how the model's predictions are explainable through analysis of individual bond importance for a biobased nylon. Overall, PolyID can aid the biobased polymer practitioner to navigate the vast number of renewable polymers to discover sustainable materials with enhanced performance.

SMiPoly: Generation of a Synthesizable Polymer Virtual Library Using Rule-Based Polymerization Reactions

Recent advances in machine learning have led to the rapid adoption of various computational methods for de novo molecular design in polymer research, including high-throughput virtual screening and inverse molecular design. In such workflows, molecular generators play an essential role in creation or sequential modification of candidate polymer structures. Machine learning-assisted molecular design has made great technical progress over the past few years. However, the difficulty of identifying synthetic routes to such designed polymers remains unresolved. To address this technical limitation, we present Small Molecules into Polymers (SMiPoly), a Python library for virtual polymer generation that implements 22 chemical rules for commonly applied polymerization reactions. For given small organic molecules to form a candidate monomer set, the SMiPoly generator conducts possible polymerization reactions to generate an exhaustive list of potentially synthesizable polymers. In this study, using 1083 readily available monomers, we generated 169,347 unique polymers forming seven different molecular types: polyolefin, polyester, polyether, polyamide, polyimide, polyurethane, and polyoxazolidone. By comparing the distribution of the virtually created polymers with approximately 16,000 real polymers synthesized so far, it was found that the coverage and novelty of the SMiPoly-generated polymers can reach 48 and 53%, respectively. Incorporating the SMiPoly library into a molecular design workflow will accelerate the process of de novo polymer synthesis by shortening the step to select synthesizable candidate polymers.

Applied machine learning as a driver for polymeric biomaterials design

Polymers are ubiquitous to almost every aspect of modern society and their use in medical products is similarly pervasive. Despite this, the diversity in commercial polymers used in medicine is stunningly low. Considerable time and resources have been extended over the years towards the development of new polymeric biomaterials which address unmet needs left by the current generation of medical-grade polymers. Machine learning (ML) presents an unprecedented opportunity in this field to bypass the need for trial-and-error synthesis, thus reducing the time and resources invested into new discoveries critical for advancing medical treatments. Current efforts pioneering applied ML in polymer design have employed combinatorial and high throughput experimental design to address data availability concerns. However, the lack of available and standardized characterization of parameters relevant to medicine, including degradation time and biocompatibility, represents a nearly insurmountable obstacle to ML-aided design of biomaterials. Herein, we identify a gap at the intersection of applied ML and biomedical polymer design, highlight current works at this junction more broadly and provide an outlook on challenges and future directions.

DeepRT: Predicting compounds presence in pathway modules and classifying into module classes using deep neural networks based on molecular properties

Metabolic pathways play a crucial role in understanding the biochemistry of organisms. In metabolic pathways, modules refer to clusters of interconnected reactions or sub-networks representing specific functional units or biological processes within the overall pathway. In pathway modules, compounds are major elements and refer to the various molecules that participate in the biochemical reactions within the pathway modules. These molecules can include substrates, intermediates and final products. Determining the presence relation of compounds and pathway modules is essential for synthesizing new molecules and predicting hidden reactions. To date, several computational methods have been proposed to address this problem. However, all methods only predict the metabolic pathways and their types, not the pathway modules. To address this issue, we proposed a novel deep learning model, DeepRT that integrates message passing neural networks (MPNNs) and transformer encoder. This combination allows DeepRT to effectively extract global and local structure information from the molecular graph. The model is designed to perform two tasks: first, determining the present relation of the compound with the pathway module, and second, predicting the relation of query compound and module classes. The proposed DeepRT model evaluated on a dataset comprising compounds and pathway modules, and it outperforms existing approaches.

Multiresolution graph transformers and wavelet positional encoding for learning long-range and hierarchical structures

Contemporary graph learning algorithms are not well-suited for large molecules since they do not consider the hierarchical interactions among the atoms, which are essential to determining the molecular properties of macromolecules. In this work, we propose Multiresolution Graph Transformers (MGT), the first graph transformer architecture that can learn to represent large molecules at multiple scales. MGT can learn to produce representations for the atoms and group them into meaningful functional groups or repeating units. We also introduce Wavelet Positional Encoding (WavePE), a new positional encoding method that can guarantee localization in both spectral and spatial domains. Our proposed model achieves competitive results on three macromolecule datasets consisting of polymers, peptides, and protein-ligand complexes, along with one drug-like molecule dataset. Significantly, our model outperforms other state-of-the-art methods and achieves chemical accuracy in estimating molecular properties (e.g., highest occupied molecular orbital, lowest unoccupied molecular orbital, and their gap) calculated by Density Functional Theory in the polymers dataset. Furthermore, the visualizations, including clustering results on macromolecules and low-dimensional spaces of their representations, demonstrate the capability of our methodology in learning to represent long-range and hierarchical structures. Our PyTorch implementation is publicly available at https://github.com/HySonLab/Multires-Graph-Transformer.

Quantum Graph Neural Network Models for Materials Search

Inspired by classical graph neural networks, we discuss a novel quantum graph neural network (QGNN) model to predict the chemical and physical properties of molecules and materials. QGNNs were investigated to predict the energy gap between the highest occupied and lowest unoccupied molecular orbitals of small organic molecules. The models utilize the equivariantly diagonalizable unitary quantum graph circuit (EDU-QGC) framework to allow discrete link features and minimize quantum circuit embedding. The results show QGNNs can achieve lower test loss compared to classical models if a similar number of trainable variables are used, and converge faster in training. This paper also provides a review of classical graph neural network models for materials research and various QGNNs.

Integrating Machine Learning in the Coarse-Grained Molecular Simulation of Polymers

Machine learning (ML) is having an increasing impact on the physical sciences, engineering, and technology and its integration into molecular simulation frameworks holds great potential to expand their scope of applicability to complex materials and facilitate fundamental knowledge and reliable property predictions, contributing to the development of efficient materials design routes. The application of ML in materials informatics in general, and polymer informatics in particular, has led to interesting results, however great untapped potential lies in the integration of ML techniques into the multiscale molecular simulation methods for the study of macromolecular systems, specifically in the context of Coarse Grained (CG) simulations. In this Perspective, we aim at presenting the pioneering recent research efforts in this direction and discussing how these new ML-based techniques can contribute to critical aspects of the development of multiscale molecular simulation methods for bulk complex chemical systems, especially polymers. Prerequisites for the implementation of such ML-integrated methods and open challenges that need to be met toward the development of general systematic ML-based coarse graining schemes for polymers are discussed.

Machine learned interatomic potential for dispersion strengthened plasma facing components

Tungsten (W) is a material of choice for the divertor material due to its high melting temperature, thermal conductivity, and sputtering threshold. However, W has a very high brittle-to-ductile transition temperature, and at fusion reactor temperatures (≥1000 K), it may undergo recrystallization and grain growth. Dispersion-strengthening W with zirconium carbide (ZrC) can improve ductility and limit grain growth, but much of the effects of the dispersoids on microstructural evolution and thermomechanical properties at high temperatures are still unknown. We present a machine learned Spectral Neighbor Analysis Potential for W-ZrC that can now be used to study these materials. In order to construct a potential suitable for large-scale atomistic simulations at fusion reactor temperatures, it is necessary to train on ab initio data generated for a diverse set of structures, chemical environments, and temperatures. Further accuracy and stability tests of the potential were achieved using objective functions for both material properties and high temperature stability. Validation of lattice parameters, surface energies, bulk moduli, and thermal expansion is confirmed on the optimized potential. Tensile tests of W/ZrC bicrystals show that although the W(110)-ZrC(111) C-terminated bicrystal has the highest ultimate tensile strength (UTS) at room temperature, observed strength decreases with increasing temperature. At 2500 K, the terminating C layer diffuses into the W, resulting in a weaker W-Zr interface. Meanwhile, the W(110)-ZrC(111) Zr-terminated bicrystal has the highest UTS at 2500 K.

Perspectives on development of biomedical polymer materials in artificial intelligence age

Polymer materials are widely used in biomedicine, chemistry and material science, whose traditional preparations are mainly based on experience, intuition and conceptual insight, having been applied to the development of many new materials, but facing great challenges due to the vast design space for biomedical polymers. So far, the best way to solve these problems is to accelerate material design through artificial intelligence, especially machine learning. Herein, this paper will introduce several successful cases, and analyze the latest progress of machine learning in the field of biomedical polymers, then discuss the opportunities of this novel method. In particular, this paper summarizes the material database, open-source determination tools, molecular generation methods and machine learning models that have been used for biopolymer synthesis and property prediction. Overall, machine learning could be more effectively deployed on the material design of biomedical polymers, and it is expected to become an extensive driving force to meet the huge demand for customized designs.

Polymer Informatics at Scale with Multitask Graph Neural Networks

Artificial intelligence-based methods are becoming increasingly effective at screening libraries of polymers down to a selection that is manageable for experimental inquiry. The vast majority of presently adopted approaches for polymer screening rely on handcrafted chemostructural features extracted from polymer repeat units-a burdensome task as polymer libraries, which approximate the polymer chemical search space, progressively grow over time. Here, we demonstrate that directly "machine learning" important features from a polymer repeat unit is a cheap and viable alternative to extracting expensive features by hand. Our approach-based on graph neural networks, multitask learning, and other advanced deep learning techniques-speeds up feature extraction by 1-2 orders of magnitude relative to presently adopted handcrafted methods without compromising model accuracy for a variety of polymer property prediction tasks. We anticipate that our approach, which unlocks the screening of truly massive polymer libraries at scale, will enable more sophisticated and large scale screening technologies in the field of polymer informatics.

Machine Learning with Enormous "Synthetic" Data Sets: Predicting Glass Transition Temperature of Polyimides Using Graph Convolutional Neural Networks

In the present work, we address the problem of utilizing machine learning (ML) methods to predict the thermal properties of polymers by establishing "structure-property" relationships. Having focused on a particular class of heterocyclic polymers, namely polyimides (PIs), we developed a graph convolutional neural network (GCNN), being one of the most promising tools for working with big data, to predict the PI glass transition temperature T g as an example of the fundamental property of polymers. To train the GCNN, we propose an original methodology based on using a "transfer learning" approach with an enormous "synthetic" data set for pretraining and a small experimental data set for its fine-tuning. The "synthetic" data set contains more than 6 million combinatorically generated repeating units of PIs and theoretical values of their T g values calculated using the well-established Askadskii's quantitative structure-property relationship (QSPR) computational scheme. Additionally, an experimental data set for 214 PIs was also collected from the literature for training, fine-tuning, and validation of the GCNN. Both "synthetic" and experimental data sets are included into a PolyAskInG database (Polymer Askadskii's Intelligent Gateway). By using the PolyAskInG database, we developed GCNN which allows estimation of T g of PI with a mean absolute error (MAE) of about 20 K, which is 1.5 times lower than in the case of Askadskii QSPR analysis (33 K). To prove the efficiency and usability of the proposed GCNN architecture and training methodology for predicting polymer properties, we also employed "transfer learning" to develop alternative GCNN pretrained on proxy-characteristics taken from the popular quantum-chemical QM9 database for small compounds and fine-tuned on an experimental T g values data set from PolyAskInG database. The obtained results indicate that pretraining of GCNN on the "synthetic" polymer data set provides MAE which is almost twice as low as that in the case of using the QM9 data set in the pretraining stage (∼41 K). Furthermore, we address the questions associated with the influence of the differences in the size of the experimental and "synthetic" data sets (so-called "reality gap" problem), as well as their chemical composition on the training quality. Our results state the overall priority of using polymer data sets for developing deep neural networks, and GCNN in particular, for efficient prediction of polymer properties. Moreover, our work opens up a challenge for the theoretically supported generation of large "synthetic" data sets of polymer properties for the training of the complex ML models. The proposed methodology is rather versatile and may be generalized for predicting other properties of different polymers and copolymers synthesized through the polycondensation reaction.

Representing Polymers as Periodic Graphs with Learned Descriptors for Accurate Polymer Property Predictions

Accurately predicting new polymers' properties with machine learning models apriori to synthesis has potential to significantly accelerate new polymers' discovery and development. However, accurately and efficiently capturing polymers' complex, periodic structures in machine learning models remains a grand challenge for the polymer cheminformatics community. Specifically, there has yet to be an ideal solution for the problems of how to capture the periodicity of polymers, as well as how to optimally develop polymer descriptors without requiring human-based feature design. In this work, we tackle these problems by utilizing a periodic polymer graph representation that accounts for polymers' periodicity and coupling it with a message-passing neural network that leverages the power of graph deep learning to automatically learn chemically relevant polymer descriptors. Remarkably, this approach achieves state-of-the-art performance on 8 out of 10 distinct polymer property prediction tasks. These results highlight the advancement in predictive capability that is possible through learning descriptors that are specifically optimized for capturing the unique chemical structure of polymers.

Graph neural networks for materials science and chemistry

Machine learning plays an increasingly important role in many areas of chemistry and materials science, being used to predict materials properties, accelerate simulations, design new structures, and predict synthesis routes of new materials. Graph neural networks (GNNs) are one of the fastest growing classes of machine learning models. They are of particular relevance for chemistry and materials science, as they directly work on a graph or structural representation of molecules and materials and therefore have full access to all relevant information required to characterize materials. In this Review, we provide an overview of the basic principles of GNNs, widely used datasets, and state-of-the-art architectures, followed by a discussion of a wide range of recent applications of GNNs in chemistry and materials science, and concluding with a road-map for the further development and application of GNNs.

A graph representation of molecular ensembles for polymer property prediction

Synthetic polymers are versatile and widely used materials. Similar to small organic molecules, a large chemical space of such materials is hypothetically accessible. Computational property prediction and virtual screening can accelerate polymer design by prioritizing candidates expected to have favorable properties. However, in contrast to organic molecules, polymers are often not well-defined single structures but an ensemble of similar molecules, which poses unique challenges to traditional chemical representations and machine learning approaches. Here, we introduce a graph representation of molecular ensembles and an associated graph neural network architecture that is tailored to polymer property prediction. We demonstrate that this approach captures critical features of polymeric materials, like chain architecture, monomer stoichiometry, and degree of polymerization, and achieves superior accuracy to off-the-shelf cheminformatics methodologies. While doing so, we built a dataset of simulated electron affinity and ionization potential values for >40k polymers with varying monomer composition, stoichiometry, and chain architecture, which may be used in the development of other tailored machine learning approaches. The dataset and machine learning models presented in this work pave the path toward new classes of algorithms for polymer informatics and, more broadly, introduce a framework for the modeling of molecular ensembles.

Accelerating amorphous polymer electrolyte screening by learning to reduce errors in molecular dynamics simulated properties

Polymer electrolytes are promising candidates for the next generation lithium-ion battery technology. Large scale screening of polymer electrolytes is hindered by the significant cost of molecular dynamics (MD) simulation in amorphous systems: the amorphous structure of polymers requires multiple, repeated sampling to reduce noise and the slow relaxation requires long simulation time for convergence. Here, we accelerate the screening with a multi-task graph neural network that learns from a large amount of noisy, unconverged, short MD data and a small number of converged, long MD data. We achieve accurate predictions of 4 different converged properties and screen a space of 6247 polymers that is orders of magnitude larger than previous computational studies. Further, we extract several design principles for polymer electrolytes and provide an open dataset for the community. Our approach could be applicable to a broad class of material discovery problems that involve the simulation of complex, amorphous materials.

Screening polymer electrolytes for batteries is extremely expensive due to the complex structures and slow dynamics. Here the authors develop a machine learning scheme to accelerate the screening and explore a space much larger than past studies.

Language models can learn complex molecular distributions

Deep generative models of molecules have grown immensely in popularity, trained on relevant datasets, these models are used to search through chemical space. The downstream utility of generative models for the inverse design of novel functional compounds, depends on their ability to learn a training distribution of molecules. The most simple example is a language model that takes the form of a recurrent neural network and generates molecules using a string representation. Since their initial use, subsequent work has shown that language models are very capable, in particular, recent research has demonstrated their utility in the low data regime. In this work, we investigate the capacity of simple language models to learn more  complex distributions of molecules. For this purpose, we introduce several challenging generative modeling tasks by compiling larger, more complex distributions of molecules and we evaluate the ability of language models on each task. The results demonstrate that language models are powerful generative models, capable of adeptly learning complex molecular distributions. Language models can accurately generate: distributions of the highest scoring penalized LogP molecules in ZINC15, multi-modal molecular distributions as well as the largest molecules in PubChem. The results highlight the limitations of some of the most popular and recent graph generative models- many of which cannot scale to these molecular distributions.

Chemistry-informed macromolecule graph representation for similarity computation, unsupervised and supervised learning

The near-infinite chemical diversity of natural and artificial macromolecules arises from the vast range of possible component monomers, linkages, and polymers topologies. This enormous variety contributes to the ubiquity and indispensability of macromolecules but hinders the development of general machine learning methods with macromolecules as input. To address this, we developed a chemistry-informed graph representation of macromolecules that enables quantifying structural similarity, and interpretable supervised learning for macromolecules. Our work enables quantitative chemistry-informed decision-making and iterative design in the macromolecular chemical space.

Machine Learning-Assisted Development of Organic Solar Cell Materials: Issues, Analyses, and Outlooks

Nonfullerene, a small molecular electron acceptor, has substantially improved the power conversion efficiency of organic photovoltaics (OPVs). However, the large structural freedom of π-conjugated polymers and molecules makes it difficult to explore with limited resources. Machine learning, which is based on rapidly growing artificial intelligence technology, is a high-throughput method to accelerate the speed of material design and process optimization; however, it suffers from limitations in terms of prediction accuracy, interpretability, data collection, and available data (particularly, experimental data). This recognition motivates the present Perspective, which focuses on utilizing the experimental data set for ML to efficiently aid OPV research. This Perspective discusses the trends in ML-OPV publications, the NFA category, and the effects of data size and explanatory variables (fingerprints or Mordred descriptors) on the prediction accuracy and explainability, which broadens the scope of ML and would be useful for the development of next-generation solar cell materials.

Decoding Interaction Patterns from the Chemical Sequence of Polymers Using Neural Networks

The relation between chemical sequences and the properties of polymers is considered using artificial neural networks with a low-dimensional bottleneck layer of neurons. These encoder-decoder architectures may compress the input information into a meaningful set of physical variables that describe the correlation between distinct types of data. In this work, neural networks were trained to translate a sequence of hydrophilic and hydrophobic segments into the effective free energy landscape of a copolymer interacting with a lipid membrane. The training data were obtained by the sampling of coarse-grained polymer conformations in a given membrane density field. Neural networks that were split into separate channels have learned to decompose the free energy into independent components that are explainable by known concepts from polymer physics. The semantic information in the hidden layers was employed to predict polymer translocation events through a membrane for a more detailed dynamic model via a transfer learning procedure. The search for minimal translocation times in the compressed chemical space underlined that nontrivial sequence motifs may lead to optimal properties.

Artificial Intelligence for Autonomous Molecular Design: A Perspective

Domain-aware artificial intelligence has been increasingly adopted in recent years to expedite molecular design in various applications, including drug design and discovery. Recent advances in areas such as physics-informed machine learning and reasoning, software engineering, high-end hardware development, and computing infrastructures are providing opportunities to build scalable and explainable AI molecular discovery systems. This could improve a design hypothesis through feedback analysis, data integration that can provide a basis for the introduction of end-to-end automation for compound discovery and optimization, and enable more intelligent searches of chemical space. Several state-of-the-art ML architectures are predominantly and independently used for predicting the properties of small molecules, their high throughput synthesis, and screening, iteratively identifying and optimizing lead therapeutic candidates. However, such deep learning and ML approaches also raise considerable conceptual, technical, scalability, and end-to-end error quantification challenges, as well as skepticism about the current AI hype to build automated tools. To this end, synergistically and intelligently using these individual components along with robust quantum physics-based molecular representation and data generation tools in a closed-loop holds enormous promise for accelerated therapeutic design to critically analyze the opportunities and challenges for their more widespread application. This article aims to identify the most recent technology and breakthrough achieved by each of the components and discusses how such autonomous AI and ML workflows can be integrated to radically accelerate the protein target or disease model-based probe design that can be iteratively validated experimentally. Taken together, this could significantly reduce the timeline for end-to-end therapeutic discovery and optimization upon the arrival of any novel zoonotic transmission event. Our article serves as a guide for medicinal, computational chemistry and biology, analytical chemistry, and the ML community to practice autonomous molecular design in precision medicine and drug discovery.

Real-time prediction of 1H and 13C chemical shifts with DFT accuracy using a 3D graph neural network

Nuclear magnetic resonance (NMR) is one of the primary techniques used to elucidate the chemical structure, bonding, stereochemistry, and conformation of organic compounds. The distinct chemical shifts in an NMR spectrum depend upon each atom's local chemical environment and are influenced by both through-bond and through-space interactions with other atoms and functional groups. The in silico prediction of NMR chemical shifts using quantum mechanical (QM) calculations is now commonplace in aiding organic structural assignment since spectra can be computed for several candidate structures and then compared with experimental values to find the best possible match. However, the computational demands of calculating multiple structural- and stereo-isomers, each of which may typically exist as an ensemble of rapidly-interconverting conformations, are expensive. Additionally, the QM predictions themselves may lack sufficient accuracy to identify a correct structure. In this work, we address both of these shortcomings by developing a rapid machine learning (ML) protocol to predict ¹H and ¹³C chemical shifts through an efficient graph neural network (GNN) using 3D structures as input. Transfer learning with experimental data is used to improve the final prediction accuracy of a model trained using QM calculations. When tested on the CHESHIRE dataset, the proposed model predicts observed ¹³C chemical shifts with comparable accuracy to the best-performing DFT functionals (1.5 ppm) in around 1/6000 of the CPU time. An automated prediction webserver and graphical interface are accessible online at http://nova.chem.colostate.edu/cascade/. We further demonstrate the model in three applications: first, we use the model to decide the correct organic structure from candidates through experimental spectra, including complex stereoisomers; second, we automatically detect and revise incorrect chemical shift assignments in a popular NMR database, the NMRShiftDB; and third, we use NMR chemical shifts as descriptors for determination of the sites of electrophilic aromatic substitution.

From quantum chemical and experimental NMR data, a 3D graph neural network, CASCADE, has been developed to predict carbon and proton chemical shifts. Stereoisomers and conformers of organic molecules can be correctly distinguished.

Data-driven algorithms for inverse design of polymers

The ever-increasing demand for novel polymers with superior properties requires a deeper understanding and exploration of the chemical space. Recently, data-driven approaches to explore the chemical space for polymer design have emerged. Among them, inverse design strategies for designing polymers with specific properties have evolved to be a significant materials informatics platform by learning hidden knowledge from materials data as well as smartly navigating the chemical space in an optimized way. In this review, we first summarize the progress in the representation of polymers, a prerequisite step for the inverse design of polymers. Then, we systematically introduce three data-driven strategies implemented for the inverse design of polymers, i.e., high-throughput virtual screening, global optimization, and generative models. Finally, we discuss the challenges and opportunities of the data-driven strategies as well as optimization algorithms employed in the inverse design of polymers.

Graph-Based Approaches for Predicting Solvation Energy in Multiple Solvents: Open Datasets and Machine Learning Models

The solvation properties of molecules, often estimated using quantum chemical simulations, are important in the synthesis of energy storage materials, drugs, and industrial chemicals. Here, we develop machine learning models of solvation energies to replace expensive quantum chemistry calculations with inexpensive-to-compute message-passing neural network models that require only the molecular graph as inputs. Our models are trained on a new database of solvation energies for 130,258 molecules taken from the QM9 dataset computed in five solvents (acetone, ethanol, acetonitrile, dimethyl sulfoxide, and water) via an implicit solvent model. Our best model achieves a mean absolute error of 0.5 kcal/mol for molecules with nine or fewer non-hydrogen atoms and 1 kcal/mol for molecules with between 10 and 14 non-hydrogen atoms. We make the entire dataset of 651,290 computed entries openly available and provide simple web and programmatic interfaces to enable others to run our solvation energy model on new molecules. This model calculates the solvation energies for molecules using only the SMILES string and also provides an estimate of whether each molecule is within the domain of applicability of our model. We envision that the dataset and models will provide the functionality needed for the rapid screening of large chemical spaces to discover improved molecules for many applications.

Bioactivity descriptors for uncharacterized chemical compounds

Chemical descriptors encode the physicochemical and structural properties of small molecules, and they are at the core of chemoinformatics. The broad release of bioactivity data has prompted enriched representations of compounds, reaching beyond chemical structures and capturing their known biological properties. Unfortunately, bioactivity descriptors are not available for most small molecules, which limits their applicability to a few thousand well characterized compounds. Here we present a collection of deep neural networks able to infer bioactivity signatures for any compound of interest, even when little or no experimental information is available for them. Our signaturizers relate to bioactivities of 25 different types (including target profiles, cellular response and clinical outcomes) and can be used as drop-in replacements for chemical descriptors in day-to-day chemoinformatics tasks. Indeed, we illustrate how inferred bioactivity signatures are useful to navigate the chemical space in a biologically relevant manner, unveiling higher-order organization in natural product collections, and to enrich mostly uncharacterized chemical libraries for activity against the drug-orphan target Snail1. Moreover, we implement a battery of signature-activity relationship (SigAR) models and show a substantial improvement in performance, with respect to chemistry-based classifiers, across a series of biophysics and physiology activity prediction benchmarks.

Perspective on integrating machine learning into computational chemistry and materials science

Machine learning (ML) methods are being used in almost every conceivable area of electronic structure theory and molecular simulation. In particular, ML has become firmly established in the construction of high-dimensional interatomic potentials. Not a day goes by without another proof of principle being published on how ML methods can represent and predict quantum mechanical properties-be they observable, such as molecular polarizabilities, or not, such as atomic charges. As ML is becoming pervasive in electronic structure theory and molecular simulation, we provide an overview of how atomistic computational modeling is being transformed by the incorporation of ML approaches. From the perspective of the practitioner in the field, we assess how common workflows to predict structure, dynamics, and spectroscopy are affected by ML. Finally, we discuss how a tighter and lasting integration of ML methods with computational chemistry and materials science can be achieved and what it will mean for research practice, software development, and postgraduate training.

XGraphBoost: Extracting Graph Neural Network-Based Features for a Better Prediction of Molecular Properties

Determining the properties of chemical molecules is essential for screening candidates similar to a specific drug. These candidate molecules are further evaluated for their target binding affinities, side effects, target missing probabilities, etc. Conventional machine learning algorithms demonstrated satisfying prediction accuracies of molecular properties. A molecule cannot be directly loaded into a machine learning model, and a set of engineered features needs to be designed and calculated from a molecule. Such hand-crafted features rely heavily on the experiences of the investigating researchers. The concept of graph neural networks (GNNs) was recently introduced to describe the chemical molecules. The features may be automatically and objectively extracted from the molecules through various types of GNNs, e.g., GCN (graph convolution network), GGNN (gated graph neural network), DMPNN (directed message passing neural network), etc. However, the training of a stable GNN model requires a huge number of training samples and a large amount of computing power, compared with the conventional machine learning strategies. This study proposed the integrated framework XGraphBoost to extract the features using a GNN and build an accurate prediction model of molecular properties using the classifier XGBoost. The proposed framework XGraphBoost fully inherits the merits of the GNN-based automatic molecular feature extraction and XGBoost-based accurate prediction performance. Both classification and regression problems were evaluated using the framework XGraphBoost. The experimental results strongly suggest that XGraphBoost may facilitate the efficient and accurate predictions of various molecular properties. The source code is freely available to academic users at https://github.com/chenxiaowei-vincent/XGraphBoost.git.

A perspective on biomass-derived biofuels: From catalyst design principles to fuel properties

The hazards to health and the environment associated with the transportation sector include smog, particulate matter, and greenhouse gas emissions. Conversion of lignocellulosic biomass into biofuels has the potential to provide significant amounts of infrastructure-compatible liquid transportation fuels that reduce those hazardous materials. However, the development of these technologies is inefficient, due to: (i) the lack of a priori fuel property consideration, (ii) poor shared vocabulary between process chemists and fuel engineers, and (iii) modern and future engines operating outside the range of traditional autoignition metrics such as octane or cetane numbers. In this perspective, we describe an approach where we follow a "fuel-property first" design methodology with a sequence of (i) identifying the desirable fuel properties for modern engines, (ii) defining molecules capable of delivering those properties, and (iii) designing catalysts and processes that can produce those molecules from a candidate feedstock in a specific conversion process. Computational techniques need to be leveraged to minimize expenses and experimental efforts on low-promise options. This concept is illustrated with current research information available for biomass conversion to fuels via catalytic fast pyrolysis and hydrotreating; outstanding challenges and research tools necessary for a successful outcome are presented.

Targeted sequence design within the coarse-grained polymer genome

The chemical design of polymers with target structural and/or functional properties represents a grand challenge in materials science. While data-driven design approaches are promising, success with polymers has been limited, largely due to limitations in data availability. Here, we demonstrate the targeted sequence design of single-chain structure in polymers by combining coarse-grained modeling, machine learning, and model optimization. Nearly 2000 unique coarse-grained polymers are simulated to construct and analyze machine learning models. We find that deep neural networks inexpensively and reliably predict structural properties with limited sequence information as input. By coupling trained ML models with sequential model-based optimization, polymer sequences are proposed to exhibit globular, swollen, or rod-like behaviors, which are verified by explicit simulations. This work highlights the promising integration of coarse-grained modeling with data-driven design and represents a necessary and crucial step toward more complex polymer design efforts.

Prediction of organic homolytic bond dissociation enthalpies at near chemical accuracy with sub-second computational cost

Bond dissociation enthalpies (BDEs) of organic molecules play a fundamental role in determining chemical reactivity and selectivity. However, BDE computations at sufficiently high levels of quantum mechanical theory require substantial computing resources. In this paper, we develop a machine learning model capable of accurately predicting BDEs for organic molecules in a fraction of a second. We perform automated density functional theory (DFT) calculations at the M06-2X/def2-TZVP level of theory for 42,577 small organic molecules, resulting in 290,664 BDEs. A graph neural network trained on a subset of these results achieves a mean absolute error of 0.58 kcal mol^-1 (vs DFT) for BDEs of unseen molecules. We further demonstrate the model on two applications: first, we rapidly and accurately predict major sites of hydrogen abstraction in the metabolism of drug-like molecules, and second, we determine the dominant molecular fragmentation pathways during soot formation.

SMILES-X: autonomous molecular compounds characterization for small datasets without descriptors

There is more and more evidence that machine learning can be successfully applied in materials science and related fields. However, datasets in these fields are often quite small (from tens to several thousands of samples). This means the most advanced machine learning techniques remain neglected, as they are considered to be applicable to big data only. Moreover, materials informatics methods often rely on human-engineered descriptors, that should be carefully chosen, or even created, to fit the physicochemical property that one intends to predict. In this article, we propose a new method that tackles both the issue of small datasets and the difficulty of developing task-specific descriptors. The SMILES-X is an autonomous pipeline for molecular compounds characterisation based on a {Embed-Encode-Attend-Predict} neural architecture with a data-specific Bayesian hyper-parameters optimisation. The only input to the architecture—the SMILES strings—are de-canonicalised in order to efficiently augment the data. One of the key features of the architecture is the attention mechanism, which enables the interpretation of output predictions without extra computational cost. The SMILES-X achieves state-of-the-art results in the inference of aqueous solubility ( RMSE ¯ test ≃ 0.57 ± 0.07 mols/L), hydration free energy ( RMSE ¯ test ≃ 0.81 ± 0.22 kcal/mol, which is ∼24.5% better than molecular dynamics simulations), and octanol/water distribution coefficient ( RMSE ¯ test ≃ 0.59 ± 0.02 for LogD at pH 7.4) of molecular compounds. The SMILES-X is intended to become an important asset in the toolkit of materials scientists and chemists. The source code for the SMILES-X is available at github.com/GLambard/SMILES-X.

Deep learning for automated classification and characterization of amorphous materials

It is difficult to quantify structure-property relationships and to identify structural features of complex materials. The characterization of amorphous materials is especially challenging because their lack of long-range order makes it difficult to define structural metrics. In this work, we apply deep learning algorithms to accurately classify amorphous materials and characterize their structural features. Specifically, we show that convolutional neural networks and message passing neural networks can classify two-dimensional liquids and liquid-cooled glasses from molecular dynamics simulations with greater than 0.98 AUC, with no a priori assumptions about local particle relationships, even when the liquids and glasses are prepared at the same inherent structure energy. Furthermore, we demonstrate that message passing neural networks surpass convolutional neural networks in this context in both accuracy and interpretability. We extract a clear interpretation of how message passing neural networks evaluate liquid and glass structures by using a self-attention mechanism. Using this interpretation, we derive three novel structural metrics that accurately characterize glass formation. The methods presented here provide a procedure to identify important structural features in materials that could be missed by standard techniques and give unique insight into how these neural networks process data.