RESUMO
Hybrid density functional calculations are essential for accurate description of electronic structure, yet their widespread use is restricted by the substantial computational cost. Here we develop DeepH-hybrid, a deep equivariant neural network method for learning the hybrid-functional Hamiltonian as a function of material structure, which circumvents the time-consuming self-consistent field iterations and enables the study of large-scale materials with hybrid-functional accuracy. Our extensive experiments demonstrate good reliability as well as effective transferability and efficiency of the method. As a notable application, DeepH-hybrid is applied to study large-supercell Moiré-twisted materials, offering the first case study on how the inclusion of exact exchange affects flat bands in magic-angle twisted bilayer graphene. The work generalizes deep-learning electronic structure methods to beyond conventional density functional theory, facilitating the development of deep-learning-based ab initio methods.
RESUMO
Deep-learning density functional theory (DFT) shows great promise to significantly accelerate material discovery and potentially revolutionize materials research. However, current research in this field primarily relies on data-driven supervised learning, making the developments of neural networks and DFT isolated from each other. In this work, we present a theoretical framework of neural-network DFT, which unifies the optimization of neural networks with the variational computation of DFT, enabling physics-informed unsupervised learning. Moreover, we develop a differential DFT code incorporated with deep-learning DFT Hamiltonian, and introduce algorithms of automatic differentiation and backpropagation into DFT, demonstrating the capability of neural-network DFT. The physics-informed neural-network architecture not only surpasses conventional approaches in accuracy and efficiency, but also offers a new paradigm for developing deep-learning DFT methods.
RESUMO
Realizing large materials models has emerged as a critical endeavor for materials research in the new era of artificial intelligence, but how to achieve this fantastic and challenging objective remains elusive. Here, we propose a feasible pathway to address this paramount pursuit by developing universal materials models of deep-learning density functional theory Hamiltonian (DeepH), enabling computational modeling of the complicated structure-property relationship of materials in general. By constructing a large materials database and substantially improving the DeepH method, we obtain a universal materials model of DeepH capable of handling diverse elemental compositions and material structures, achieving remarkable accuracy in predicting material properties. We further showcase a promising application of fine-tuning universal materials models for enhancing specific materials models. This work not only demonstrates the concept of DeepH's universal materials model but also lays the groundwork for developing large materials models, opening up significant opportunities for advancing artificial intelligence-driven materials discovery.
RESUMO
Calculating perturbation response properties of materials from first principles provides a vital link between theory and experiment, but is bottlenecked by the high computational cost. Here, a general framework is proposed to perform density functional perturbation theory (DFPT) calculations by neural networks, greatly improving the computational efficiency. Automatic differentiation is applied on neural networks, facilitating accurate computation of derivatives. High efficiency and good accuracy of the approach are demonstrated by studying electron-phonon coupling and related physical quantities. This work brings deep-learning density functional theory and DFPT into a unified framework, creating opportunities for developing ab initio artificial intelligence.
RESUMO
(1) Background: The three sloe bugs, Dolycoris baccarum, Dolycoris indicus, and Dolycoris penicillatus, are found in the Chinese mainland and are morphologically similar. The species boundaries and phylogenetic relationships of the three species remain uncertain; (2) Methods: In this study, we generated multiple mitochondrial genomes (mitogenomes) for each of the three species and conducted comparative mitogenomic analysis, species delimitation, and phylogenetic analysis based on these data; (3) Results: Mitogenomes of the three Dolycoris species are conserved in nucleotide composition, gene arrangement, and codon usage. All protein-coding genes (PCGs) were found to be under purifying selection, and the ND4 evolved at the fastest rate. Most species delimitation analyses based on the COI gene and the concatenated 13 PCGs retrieved three operational taxonomic units (OTUs), which corresponded well with the three Dolycoris species identified based on morphological characters. A clear-cut barcode gap was discovered between the interspecific and intraspecific genetic distances of the three Dolycoris species. Phylogenetic analyses strongly supported the monophyly of Dolycoris, with interspecific relationship inferred as (D. indicus + (D. baccarum + D. penicillatus)); (4) Conclusions: Our study provides the first insight into the species boundaries and phylogenetic relationships of the three Dolycoris species distributed across the Chinese mainland.
RESUMO
The redbanded stink bug, Piezodorus guildinii (Westwood, 1837), is a highly destructive soybean pest native to the Neotropical Region. In the past 60 yr, P. guildinii has been observed to expand its distribution in North and South America, causing significant soybean yield losses. In order to predict the future distribution direction of P. guildinii and create an effective pest control strategy, we projected the potential global distribution of P. guildinii using 2 different emission scenarios, Shared Socioeconomic Pathways 126 and 585, and 3 Earth system models, with the maximum entropy niche model (MaxEnt). Then, the predicted distribution areas of P. guildinii were jointly analyzed with the main soybean-producing areas to assess the impact for different soybean region. Our results showed that temperature is the main environmental factor limiting the distribution of P. guildinii. Under present climate conditions, all continents except Antarctica have suitable habitat for P. guildinii. These suitable habitats overlap with approximately 45.11% of the total global cultivated soybean areas. Moreover, P. guildinii was predicted to expand its range in the future, particularly into higher latitudes in the Northern hemisphere. Countries, in particular the United States, where soybean is widely available, would face a management challenge under global warming. In addition, China and India are also high-risk countries that may be invaded and should take strict quarantine measures. The maps of projected distribution produced in this study may prove useful in the future management of P. guildinii and the containment of its disruptive effects.
Assuntos
Hemípteros , Heterópteros , Animais , Glycine max , Mudança Climática , China , ÍndiaRESUMO
Ab initio studies of magnetic superstructures are indispensable to research on emergent quantum materials, but are currently bottlenecked by the formidable computational cost. Here, to break this bottleneck, we have developed a deep equivariant neural network framework to represent the density functional theory Hamiltonian of magnetic materials for efficient electronic-structure calculation. A neural network architecture incorporating a priori knowledge of fundamental physical principles, especially the nearsightedness principle and the equivariance requirements of Euclidean and time-reversal symmetries ([Formula: see text]), is designed, which is critical to capture the subtle magnetic effects. Systematic experiments on spin-spiral, nanotube and moiré magnets were performed, making the challenging study of magnetic skyrmions feasible.
RESUMO
Coreoidea (Insecta: Hemiptera: Heteroptera) is a widely distributed and agriculturally important bugs. However, the phylogeny of Coreoidea lacked consensus on higher-level relationships and several studies by comparative morphological characters and molecular data suggested the non-monophyly of two families: Coreidae and Alydidae. The mitochondrial genome (mitogenome) has long been thought to be a significant marker to understand phylogenetic relationships, but the mitogenome in Alydidae is scarce to date. In the present study, we gathered the mitogenomes of 28 species from four families of Coreoidea excluding Hyocephalidae (Alydidae, Coreidae, Rhopalidae, and Stenocephalidae), including four newly sequenced mitogenomes of Alydidae, and conducted mitogenomic organization and phylogenetic studies. We used maximum likelihood and Bayesian inference methods to infer the higher-level phylogeny from the perspective of mitogenomes, primarily to investigate the phylogenetic relationship betweeen Coreidae and Alydidae. We add evidence that neither Alydidae nor Coreidae are monophyletic based on mitogenomes. Newly sequenced mitogenomes of Alydidae have traditional gene structure and gene rearrangement was not found. Alydinae was always recovered as closely related to Pseudophloeinae of the coreid subfamily with high support. The placement of the coreid subfamily Hydarinae and alydid subfamily Micrelytrinae are unstable depending on approach used. In terms of the length and nucleotide composition of the protein coding genes in mitogenomes, Pseudophloeinae and Hydarinae of coreid were more similar to Alydidae. The unsettled classification issues of Coreidae and Alydidae by mitogenomes were demonstrated in this work, indicating that further study is needed.