Artificial intelligence in perinatal mental health research: A scoping review.

The intersection of Artificial Intelligence (AI) and perinatal mental health research presents promising avenues, yet uncovers significant challenges for innovation. This review explicitly focuses on this multidisciplinary field and undertakes a comprehensive exploration of existing research therein. Through a scoping review guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, we searched relevant literature spanning a decade (2013-2023) and selected fourteen studies for our analysis. We first provide an overview of the main AI techniques and their development, including traditional methods across different categories, as well as recent emerging methods in the field. Then, through our analysis of the literature, we summarize the predominant AI and ML techniques adopted and their applications in perinatal mental health studies, such as identifying risk factors, predicting perinatal mental health disorders, voice assistants, and Q&A chatbots. We also discuss existing limitations and potential challenges that hinder AI technologies from improving perinatal mental health outcomes, and suggest several promising directions for future research to meet real needs in the field and facilitate the translation of research into clinical settings.

TransAC4C-a novel interpretable architecture for multi-species identification of N4-acetylcytidine sites in RNA with single-base resolution.

N4-acetylcytidine (ac4C) is a modification found in ribonucleic acid (RNA) related to diseases. Expensive and labor-intensive methods hindered the exploration of ac4C mechanisms and the development of specific anti-ac4C drugs. Therefore, an advanced prediction model for ac4C in RNA is urgently needed. Despite the construction of various prediction models, several limitations exist: (1) insufficient resolution at base level for ac4C sites; (2) lack of information on species other than Homo sapiens; (3) lack of information on RNA other than mRNA; and (4) lack of interpretation for each prediction. In light of these limitations, we have reconstructed the previous benchmark dataset and introduced a new dataset including balanced RNA sequences from multiple species and RNA types, while also providing base-level resolution for ac4C sites. Additionally, we have proposed a novel transformer-based architecture and pipeline for predicting ac4C sites, allowing for highly accurate predictions, visually interpretable results and no restrictions on the length of input RNA sequences. Statistically, our work has improved the accuracy of predicting specific ac4C sites in multiple species from less than 40% to around 85%, achieving a high AUC > 0.9. These results significantly surpass the performance of all existing models.

Strong Local Polarization Fluctuations Enabled High Electrostatic Energy Storage in Pb-Free Relaxors.

Electrostatic energy-storage ceramic capacitors are essential components of modern electrified power systems. However, improving their energy-storage density while maintaining high efficiency to facilitate cutting-edge miniaturized and integrated applications remains an ongoing challenge. Herein, we report a record-high energy-storage density of 20.3 J cm-3 together with a high efficiency of 89.3% achieved by constructing a relaxor ferroelectric state with strongly enhanced local polarization fluctuations. This is realized by incorporating highly polarizable, heterovalent, and large-sized Zn and Nb ions into a Bi0.5Na0.5TiO3-BaTiO3 ferroelectric matrix with very strong tetragonal distortion. Element-specific local structure analysis revealed that the foreign ions strengthen the magnitude of the unit-cell polarization vectors while simultaneously reducing their orientation anisotropy and forming strong fluctuations in both magnitude and orientation within 1-3 nm polar clusters. This leads to a particularly high polarization variation (ΔP) of 72 µC cm-2, low hysteresis, and a high effective polarization coefficient at a high breakdown strength of 80 kV mm-1. This work has surpassed the current energy density limit of 20 J cm-3 in bulk Pb-free ceramics and has demonstrated that controlling the local structure via the chemical composition design can open up new possibilities for exploring relaxors with high energy-storage performance.

Role of dietary nutrients and metabolism in colorectal cancer.

Colorectal cancer (CRC) is one of the most common malignancies and the leading causes of cancer related deaths worldwide. The development of CRC is driven by a combination of genetic and environmental factors. There is growing evidence that changes in dietary nutrition may modulate the CRC risk, and protective effects on the risk of developing CRC have been advocated for specific nutrients such as glucose, amino acids, lipid, vitamins, micronutrients and prebiotics. Metabolic crosstalk between tumor cells, tumor microenvironment components and intestinal flora further promote proliferation, invasion and metastasis of CRC cells and leads to treatment resistance. This review summarizes the research progress on CRC prevention, pathogenesis, and treatment by dietary supplementation or deficiency of glucose, amino acids, lipids, vitamins, micronutri-ents, and prebiotics, respectively. The roles played by different nutrients and dietary crosstalk in the tumor microenvironment and metabolism are discussed, and nutritional modulation is inspired to be beneficial in the prevention and treatment of CRC.

A Study on the Field Emission Characteristics of High-Quality Wrinkled Multilayer Graphene Cathodes.

Field emission (FE) necessitates cathode materials with low work function and high thermal and electrical conductivity and stability. To meet these requirements, we developed FE cathodes based on high-quality wrinkled multilayer graphene (MLG) prepared using the bubble-assisted chemical vapor deposition (B-CVD) method and investigated their emission characteristics. The result showed that MLG cathodes prepared using the spin-coating method exhibited a high field emission current density (~7.9 mA/cm2), indicating the excellent intrinsic emission performance of the MLG. However, the weak adhesion between the MLG and the substrate led to the poor stability of the cathode. Screen printing was employed to prepare the cathode to improve stability, and the influence of a silver buffer layer was explored on the cathode's performance. The results demonstrated that these cathodes exhibited better emission stability, and the silver buffer layer further enhanced the comprehensive field emission performance. The optimized cathode possesses low turn-on field strength (~1.5 V/µm), low threshold field strength (~2.65 V/µm), high current density (~10.5 mA/cm2), and good emission uniformity. Moreover, the cathode also exhibits excellent emission stability, with a current fluctuation of only 6.28% during a 4-h test at 1530 V.

IGCNSDA: unraveling disease-associated snoRNAs with an interpretable graph convolutional network.

Accurately delineating the connection between short nucleolar RNA (snoRNA) and disease is crucial for advancing disease detection and treatment. While traditional biological experimental methods are effective, they are labor-intensive, costly and lack scalability. With the ongoing progress in computer technology, an increasing number of deep learning techniques are being employed to predict snoRNA-disease associations. Nevertheless, the majority of these methods are black-box models, lacking interpretability and the capability to elucidate the snoRNA-disease association mechanism. In this study, we introduce IGCNSDA, an innovative and interpretable graph convolutional network (GCN) approach tailored for the efficient inference of snoRNA-disease associations. IGCNSDA leverages the GCN framework to extract node feature representations of snoRNAs and diseases from the bipartite snoRNA-disease graph. SnoRNAs with high similarity are more likely to be linked to analogous diseases, and vice versa. To facilitate this process, we introduce a subgraph generation algorithm that effectively groups similar snoRNAs and their associated diseases into cohesive subgraphs. Subsequently, we aggregate information from neighboring nodes within these subgraphs, iteratively updating the embeddings of snoRNAs and diseases. The experimental results demonstrate that IGCNSDA outperforms the most recent, highly relevant methods. Additionally, our interpretability analysis provides compelling evidence that IGCNSDA adeptly captures the underlying similarity between snoRNAs and diseases, thus affording researchers enhanced insights into the snoRNA-disease association mechanism. Furthermore, we present illustrative case studies that demonstrate the utility of IGCNSDA as a valuable tool for efficiently predicting potential snoRNA-disease associations. The dataset and source code for IGCNSDA are openly accessible at: https://github.com/altriavin/IGCNSDA.

Conventional dendritic cells 2, the targeted antigen-presenting-cell, induces enhanced type 1 immune responses in mice immunized with CVC1302 in oil formulation.

Multifunctional CD4+ T helper 1 (Th1) cells, producing IFN-γ, TNF-α and IL-2, define a correlate of vaccine-mediated protection against intracellular infection. In our previous study, we found that CVC1302 in oil formulation promoted the differentiation of IFN-γ+/TNF-α+/IL-2+Th1 cells. In order to extend the application of CVC1302 in oil formulation, this study aimed to elucidate the mechanism of action in improving the Th1 immune response. Considering the signals required for the differentiation of CD4+ T cells to Th1 cells, we detected the distribution of innate immune cells and the model antigen OVA-FITC in lymph node (LN), as well as the quantity of cytokines produced by the innate immune cells. The results of these experiments show that, cDC2 and OVA-FITC localized to interfollicular region (IFR) of the draining lymph nodes, inflammatory monocytes localized to both IFR and T cell zone, which mainly infiltrate from the blood. In this inflammatory niche within LN, CD4+ T cells were attracted into IFR by CXCL10, secreted by inflammatory monocytes, then activated by cDC2, secreting IL-12. Above all, CVC1302 in oil formulation, on the one hand, targeted antigen and inflammatory monocytes into the LN IFR in order to attract CD4+ T cells, on the other hand, targeted cDC2 to produce IL-12 in order to promote optimal Th1 differentiation. The new finding will provide a blueprint for application of immunopotentiators in optimal formulations.

Electroencephalography (EEG) based epilepsy diagnosis via multiple feature space fusion using shared hidden space-driven multi-view learning.

Epilepsy is a chronic, non-communicable disease caused by paroxysmal abnormal synchronized electrical activity of brain neurons, and is one of the most common neurological diseases worldwide. Electroencephalography (EEG) is currently a crucial tool for epilepsy diagnosis. With the development of artificial intelligence, multi-view learning-based EEG analysis has become an important method for automatic epilepsy recognition because EEG contains difficult types of features such as time-frequency features, frequency-domain features and time-domain features. However, current multi-view learning still faces some challenges, such as the difference between samples of the same class from different views is greater than the difference between samples of different classes from the same view. In view of this, in this study, we propose a shared hidden space-driven multi-view learning algorithm. The algorithm uses kernel density estimation to construct a shared hidden space and combines the shared hidden space with the original space to obtain an expanded space for multi-view learning. By constructing the expanded space and utilizing the information of both the shared hidden space and the original space for learning, the relevant information of samples within and across views can thereby be fully utilized. Experimental results on a dataset of epilepsy provided by the University of Bonn show that the proposed algorithm has promising performance, with an average classification accuracy value of 0.9787, which achieves at least 4% improvement compared to single-view methods.

Multi-omics and Multi-VOIs to predict esophageal fistula in esophageal cancer patients treated with radiotherapy.

OBJECTIVE: This study aimed to develop a prediction model for esophageal fistula (EF) in esophageal cancer (EC) patients treated with intensity-modulated radiation therapy (IMRT), by integrating multi-omics features from multiple volumes of interest (VOIs). METHODS: We retrospectively analyzed pretreatment planning computed tomographic (CT) images, three-dimensional dose distributions, and clinical factors of 287 EC patients. Nine groups of features from different combination of omics [Radiomics (R), Dosiomics (D), and RD (the combination of R and D)], and VOIs [esophagus (ESO), gross tumor volume (GTV), and EG (the combination of ESO and GTV)] were extracted and separately selected by unsupervised (analysis of variance (ANOVA) and Pearson correlation test) and supervised (Student T test) approaches. The final model performance was evaluated using five metrics: average area under the receiver-operator-characteristics curve (AUC), accuracy, precision, recall, and F1 score. RESULTS: For multi-omics using RD features, the model performance in EG model shows: AUC, 0.817 ± 0.031; 95% CI 0.805, 0.825; p < 0.001, which is better than single VOI (ESO or GTV). CONCLUSION: Integrating multi-omics features from multi-VOIs enables better prediction of EF in EC patients treated with IMRT. The incorporation of dosiomics features can enhance the model performance of the prediction.

Evaluating the Field Emission Properties of N-Type Black Silicon Cold Cathodes Based on a Three-Dimensional Model.

Black silicon (BS), a nanostructured silicon surface containing highly roughened surface morphology, has recently emerged as a promising candidate for field emission (FE) cathodes in novel electron sources due to its huge number of sharp tips with ease of large-scale fabrication and controllable geometrical shapes. However, evaluating the FE performance of BS-based nanostructures with high accuracy is still a challenge due to the increasing complexity in the surface morphology. Here, we demonstrate a 3D modeling methodology to fully characterize highly disordered BS-based field emitters randomly distributed on a roughened nonflat surface. We fabricated BS cathode samples with different morphological features to demonstrate the validity of this method. We utilize parametrized scanning electron microscopy images that provide high-precision morphology details, successfully describing the electric field distribution in field emitters and linking the theoretical analysis with the measured FE property of the complex nanostructures with high precision. The 3D model developed here reveals a relationship between the field emission performance and the density of the cones, successfully reproducing the classical relationship between current density J and electric field E (J-E curve). The proposed modeling approach is expected to offer a powerful tool to accurately describe the field emission properties of large-scale, disordered nano cold cathodes, thus serving as a guide for the design and application of BS as a field electron emission material.

Chemical Framework to Design Linear-like Relaxors toward Capacitive Energy Storage.

ABO3-type perovskite relaxor ferroelectrics (RFEs) have emerged as the preferred option for dielectric capacitive energy storage. However, the compositional design of RFEs with high energy density and efficiency poses significant challenges owing to the vast compositional space and the absence of general rules. Here, we present an atomic-level chemical framework that captures inherent characteristics in terms of radius and ferroelectric activity of ions. By categorizing A/B-site ions as host framework, rattling, ferroelectrically active, and blocking ions and assembling these four types of ions with specific criteria, linear-like relaxors with weak locally correlated and highly extendable unit-cell polarization vectors can be constructed. As example, we demonstrate two new compositions of Bi0.5K0.5TiO3-based and BaTiO3-based relaxors, showing extremely high recoverable energy densities of 17.3 and 12.1 J cm-3, respectively, both with a high efficiency of about 90%. Further, the role of different types of ions in forming heterogeneous polar structures is identified through element-specific local structure analysis using neutron total scattering combined with reverse Monte Carlo modeling. Our work not only opens up new avenues toward rational compositional design of high energy storage performance lead-free RFEs but also sheds light on atomic-level manipulation of functional properties in compositionally complex ferroelectrics.

Amorphous Oxyhalide Matters for Achieving Lithium Superionic Conduction.

The recently surged halide-based solid electrolytes (SEs) are great candidates for high-performance all-solid-state batteries (ASSBs), due to their decent ionic conductivity, wide electrochemical stability window, and good compatibility with high-voltage oxide cathodes. In contrast to the crystalline phases in halide SEs, amorphous components are rarely understood but play an important role in Li-ion conduction. Here, we reveal that the presence of amorphous component is common in halide-based SEs that are prepared via mechanochemical method. The fast Li-ion migration is found to be associated with the local chemistry of the amorphous proportion. Taking Zr-based halide SEs as an example, the amorphization process can be regulated by incorporating O, resulting in the formation of corner-sharing Zr-O/Cl polyhedrons. This structural configuration has been confirmed through X-ray absorption spectroscopy, pair distribution function analyses, and Reverse Monte Carlo modeling. The unique structure significantly reduces the energy barriers for Li-ion transport. As a result, an enhanced ionic conductivity of (1.35 ± 0.07) × 10-3 S cm-1 at 25 °C can be achieved for amorphous Li3ZrCl4O1.5. In addition to the improved ionic conductivity, amorphization of Zr-based halide SEs via incorporation of O leads to good mechanical deformability and promising electrochemical performance. These findings provide deep insights into the rational design of desirable halide SEs for high-performance ASSBs.

Analysis of CVC1302-Mediated Enhancement of Monocyte Recruitment in Inducing Immune Responses.

Monocytes (Mos) are believed to play important roles during the generation of immune response. In our previous study, CVC1302, a complex of PRRs agonists, was demonstrated to recruit Mo into lymph nodes (LNs) in order to present antigen and secret chemokines (CXCL9 and CXCL10), which attracted antigen-specific CD4+ T cells. As it is known that Mos in mice are divided into two main Mo subsets (Ly6C+ Mo and Ly6C- Mo), we aimed to clarify the CVC1302-recruiting Mo subset and functions in the establishment of immunity. In this study, we found that CVC1302 attracted both Ly6C+ Mo and Ly6C- Mo into draining LNs, which infiltrated from different origins, injection muscles and high endothelial venule (HEV), respectively. We also found that the numbers of OVA+ Ly6C+ Mo in the draining LNs were significantly higher compared with OVA+ Ly6C- Mo. However, the levels of CXCL9 and CXCL10 produced by Ly6C- Mo were significantly higher than Ly6C+ Mo, which plays important roles in attracting antigen-specific CD4+ T cells. Under the analysis of their functions in initiating immune responses, we found that the ability of the Ly6C+ monocyte was mainly capturing and presenting antigens, otherwise; the ability of the Ly6C- monocyte was mainly secreting CXCL9 and CXCL10, which attracted antigen-specific CD4+ T cells through CXCR3. These results will provide new insights into the development of new immunopotentiators and vaccines.

Multi-omics fusion with soft labeling for enhanced prediction of distant metastasis in nasopharyngeal carcinoma patients after radiotherapy.

Omics fusion has emerged as a crucial preprocessing approach in medical image processing, significantly assisting several studies. One of the challenges encountered in integrating omics data is the unpredictability arising from disparities in data sources and medical imaging equipment. Due to these differences, the distribution of omics futures exhibits spatial heterogeneity, diminishing their capacity to enhance subsequent tasks. To overcome this challenge and facilitate the integration of their joint application to specific medical objectives, this study aims to develop a fusion methodology for nasopharyngeal carcinoma (NPC) distant metastasis prediction to mitigate the disparities inherent in omics data. The multi-kernel late-fusion method can reduce the impact of these differences by mapping the features using the most suiTable single-kernel function and then combining them in a high-dimensional space that can effectively represent the data. The proposed approach in this study employs a distinctive framework incorporating a label-softening technique alongside a multi-kernel-based Radial basis function (RBF) neural network to address these limitations. An efficient representation of the data may be achieved by utilizing the multi-kernel to map the inherent features and then merging them in a space with many dimensions. However, the inflexibility of label fitting poses a constraint on using multi-kernel late-fusion methods in complex NPC datasets, hence affecting the efficacy of general classifiers in dealing with high-dimensional characteristics. The label softening increases the disparity between the two cohorts, providing a more flexible structure for allocating labels. The proposed model is evaluated on multi-omics datasets, and the results demonstrate its strength and effectiveness in predicting distant metastasis of NPC patients.

Frequency Domain Channel-Wise Attack to CNN Classifiers in Motor Imagery Brain-Computer Interfaces.

OBJECTIVE: Convolutional neural network (CNN), a classical structure in deep learning, has been commonly deployed in the motor imagery brain-computer interface (MIBCI). Many methods have been proposed to evaluate the vulnerability of such CNN models, primarily by attacking them using direct temporal perturbations. In this work, we propose a novel attacking approach based on perturbations in the frequency domain instead. METHODS: For a given natural MI trial in the frequency domain, the proposed approach, called frequency domain channel-wise attack (FDCA), generates perturbations at each channel one after another to fool the CNN classifiers. The advances of this strategy are two-fold. First, instead of focusing on the temporal domain, perturbations are generated in the frequency domain where discriminative patterns can be extracted for motor imagery (MI) classification tasks. Second, the perturbing optimization is performed based on differential evolution algorithm in a black-box scenario where detailed model knowledge is not required. RESULTS: Experimental results demonstrate the effectiveness of the proposed FDCA which achieves a significantly higher success rate than the baselines and existing methods in attacking three major CNN classifiers on four public MI benchmarks. CONCLUSION: Perturbations generated in the frequency domain yield highly competitive results in attacking MIBCI deployed by CNN models even in a black-box setting, where the model information is well-protected. SIGNIFICANCE: To our best knowledge, existing MIBCI attack approaches are all gradient-based methods and require details about the victim model, e.g., the parameters and objective function. We provide a more flexible strategy that does not require model details but still produces an effective attack outcome.

A novel clinical model for risk prediction and stratification of new-onset diabetes mellitus after distal pancreatectomy.

Background: The incidence of new-onset diabetes mellitus (NODM) after distal pancreatectomy (DP) remains high. Few studies have focused on NODM in patients with pancreatic benign or low-grade malignant lesions (PBLML). This study aimed to develop and validate an effective clinical model for risk prediction and stratification of NODM after DP in patients with PBLML. Methods: A follow-up survey was conducted to investigate NODM in patients without preoperative DM who underwent DP. Four hundred and forty-eight patients from Peking Union Medical College Hospital (PUMCH) and 178 from Guangdong Provincial People's Hospital (GDPH) met the inclusion criteria. They constituted the training cohort and the validation cohort, respectively. Univariate and multivariate Cox regression, as well as least absolute shrinkage and selection operator (LASSO) analyses, were used to identify the independent risk factors. The nomogram was constructed and verified. Concordance index (C-index), receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis (DCA) were applied to assess its predictive performance and clinical utility. Accordingly, the optimal cut-off point was determined by maximally selected rank statistics method, and the cumulative risk curves for the high- and low-risk populations were plotted to evaluate the discrimination ability of the nomogram. Results: The median follow-up duration was 42.8 months in the PUMCH cohort and 42.9 months in the GDPH cohort. The postoperative cumulative 5-year incidences of DM were 29.1% and 22.1%, respectively. Age, body mass index (BMI), length of pancreatic resection, intraoperative blood loss, and concomitant splenectomy were significant risk factors. The nomogram demonstrated significant predictive utility for post-pancreatectomy DM. The C-indexes of the nomogram were 0.739 and 0.719 in the training and validation cohorts, respectively. ROC curves demonstrated the predictive accuracy of the nomogram, and the calibration curves revealed that prediction results were in general agreement with the actual results. The considerable clinical applicability of the nomogram was certified by DCA. The optimal cut-off point for risk prediction value was 2.88, and the cumulative risk curves of each cohort showed significant differences between the high- and low-risk groups. Conclusions: The nomogram could predict and identify the NODM risk population, and provide guidance to physicians in monitoring and controlling blood glucose levels in PBLML patients after DP.

Host extracellular vesicles confer cytosolic access to systemic LPS licensing non-canonical inflammasome sensing and pyroptosis.

Intracellular surveillance for systemic microbial components during homeostasis and infections governs host physiology and immunity. However, a long-standing question is how circulating microbial ligands become accessible to intracellular receptors. Here we show a role for host-derived extracellular vesicles (EVs) in this process; human and murine plasma-derived and cell culture-derived EVs have an intrinsic capacity to bind bacterial lipopolysaccharide (LPS). Remarkably, circulating host EVs capture blood-borne LPS in vivo, and the LPS-laden EVs confer cytosolic access for LPS, triggering non-canonical inflammasome activation of gasdermin D and pyroptosis. Mechanistically, the interaction between the lipid bilayer of EVs and the lipid A of LPS underlies EV capture of LPS, and the intracellular transfer of LPS by EVs is mediated by CD14. Overall, this study demonstrates that EVs capture and escort systemic LPS to the cytosol licensing inflammasome responses, uncovering EVs as a previously unrecognized link between systemic microbial ligands and intracellular surveillance.

Increases in Cellular Immune Responses Due to Positive Effect of CVC1302-Induced Lysosomal Escape in Mice.

This study found a higher percentage of CD8+ T cells in piglets immunized with a CVC1302-adjuvanted inactivated foot-and-mouth disease virus (FMDV) vaccine. We wondered whether the CVC1302-adjuvanted inactivated FMDV vaccine promoted cellular immunity by promoting the antigen cross-presentation efficiency of ovalbumin (OVA) through dendritic cells (DCs), mainly via cytosolic pathways. This was demonstrated by the enhanced levels of lysosomal escape of OVA in the DCs loaded with OVA and CVC1302. The higher levels of ROS and significantly enhanced elevated lysosomal pH levels in the DCs facilitated the lysosomal escape of OVA. Significantly enhanced CTL activity levels was observed in the mice immunized with OVA-CVC1302. Overall, CVC1302 increased the cross-presentation of exogenous antigens and the cross-priming of CD8+ T cells by alkalizing the lysosomal pH and facilitating the lysosomal escape of antigens. These studies shed new light on the development of immunopotentiators to improve cellular immunity induced by vaccines.

A comprehensive review and evaluation of graph neural networks for non-coding RNA and complex disease associations.

Non-coding RNAs (ncRNAs) play a critical role in the occurrence and development of numerous human diseases. Consequently, studying the associations between ncRNAs and diseases has garnered significant attention from researchers in recent years. Various computational methods have been proposed to explore ncRNA-disease relationships, with Graph Neural Network (GNN) emerging as a state-of-the-art approach for ncRNA-disease association prediction. In this survey, we present a comprehensive review of GNN-based models for ncRNA-disease associations. Firstly, we provide a detailed introduction to ncRNAs and GNNs. Next, we delve into the motivations behind adopting GNNs for predicting ncRNA-disease associations, focusing on data structure, high-order connectivity in graphs and sparse supervision signals. Subsequently, we analyze the challenges associated with using GNNs in predicting ncRNA-disease associations, covering graph construction, feature propagation and aggregation, and model optimization. We then present a detailed summary and performance evaluation of existing GNN-based models in the context of ncRNA-disease associations. Lastly, we explore potential future research directions in this rapidly evolving field. This survey serves as a valuable resource for researchers interested in leveraging GNNs to uncover the complex relationships between ncRNAs and diseases.

Local Chemical Clustering Enabled Ultrahigh Capacitive Energy Storage in Pb-Free Relaxors.

Designing Pb-free relaxors with both a high capacitive energy density (Wrec) and high storage efficiency (η) remains a remarkable challenge for cutting-edge pulsed power technologies. Local compositional heterogeneity is crucial for achieving complex polar structure in solid solution relaxors, but its role in optimizing energy storage properties is often overlooked. Here, we report that an exceptionally high Wrec of 15.2 J cm-3 along with an ultrahigh η of 91% can be achieved through designing local chemical clustering in Bi0.5Na0.5TiO3-BaTiO3-based relaxors. A three-dimensional atomistic model derived from neutron/X-ray total scattering combined with reverse Monte Carlo method reveals the presence of subnanometer scale clustering of Bi, Na, and Ba, which host differentiated polar displacements, and confirming the prediction by density functional theory calculations. This leads to a polar state with small polar clusters and strong length and direction fluctuations in unit-cell polar vectors, thus manifesting improved high-field polarizability, steadily reduced hysteresis, and high breakdown strength macroscopically. The favorable polar structure features also result in a unique field-increased η, excellent stability, and superior discharge capacity. Our work demonstrates that the hidden local chemical order exerts a significant impact on the polarization characteristic of relaxors, and can be exploited for accessing superior energy storage performance.