Domain Fusion Research Articles

Meta-learning-based fault diagnosis method for rolling bearings under cross working-conditions

Abstract Problems such as domain shift and insufficient sample quantity may occur when conducting fault diagnosis under cross-working conditions, decreasing the accuracy and generalization of deep learning algorithms. In this paper, a fault diagnosis framework based on meta-learning and a dual-channel classification feature fusion network combines the advantages of meta-learning and dual-channel classification feature fusion network to enhance the framework's performance in cross-working condition diagnosis and few-shot learning. Firstly, dual-channel network is used to extract the classification features of different domains, and the features are fused. Then, training is conducted under the meta-learning strategy to acquire prior knowledge for fast model learning under cross-working conditions and solve the problem of a few shots. Finally, two public rolling bearing data sets are used to demonstrate the efficacy of the proposed method across different operational conditions. Before that, the appropriate sample length and fusion domain were selected through experiments. The proposed method also has good fault diagnosis accuracy in cross-device tasks. The experimental results verify the effective classification capability and robustness of the proposed method. Furthermore, the superiority of this method over other meta-learning approaches is evident when compared.

A hydraulic motor fault diagnosis method based on weighted multi-channel information fusion

Abstract In response to the limitations of the existing single-sensor hydraulic motor fault diagnosis model, which includes significant fluctuations in fault identification accuracy, low data utilization, poor reliability, and insufficient generalization ability under variable working conditions, a novel hydraulic motor fault diagnosis method based on weighted fusion of multi-channel data and migration learning is proposed. Firstly, in order to fully extract the information content in the multi-channel data set of the hydraulic motor, a multi channel-fusion method based on information entropy weighting is proposed. The information entropy method is employed to calculate the fusion weight of each channel of data, and the sampled data of each channel is weighted and fused. Subsequently, the fusion data from the source domain is employed to pre-train the depth transmission model, with the model parameters obtained from this pre-training serving as the initialization parameters for the target domain model. Subsequently, the parameters of the target domain model's feature extractor are fixed, and the parameters of its classifier are fine-tuned using the target domain's fusion data. The distance between the source and target domains is reduced by incorporating an attention mechanism and constructing a loss function. The migration from the source domain to the target domain is achieved, which enables the classifier to adapt to the novel target sample recognition task. Ultimately, the experimental results of hydraulic motor migration diagnosis under variable operating conditions demonstrate that the proposed method is an efficacious approach for hydraulic motor fault diagnosis. In comparison to conventional models such as
CNN and GRU, the proposed method exhibits superior migration diagnosis accuracy and enhanced generalization and robustness under variable operating conditions.

Albumin binding domain fusion improved the therapeutic efficacy of Inhibitor of Differentiation-2 protein in colitis mice

Albumin binding domain fusion improved the therapeutic efficacy of Inhibitor of Differentiation-2 protein in colitis mice

The heterologous expression of novel recombinant protein composed of HN and F moieties of Newcastle disease virus and immunogenicity evaluation in mouse model.

The rapid spread of Newcastle disease (ND), driven by extensive commercial exchange in the poultry industry, necessitates urgent preventive measures. Although effective vaccines against the Newcastle disease virus (NDV) have been used since 1940, recent outbreaks and the limitations of current vaccines highlight the need for improved solutions. Advances in synthetic biology, reverse vaccinology, molecular biology, and recombinant DNA technology over the past 20 years have led to the development of recombinant vaccines, which offer enhanced protection and broader immunogenic coverage against NDV. This study aimed to express the immunogenic domains of Hemagglutinin Neuraminidase (HN) and Fusion (F) glycoproteins, linked to the heat-labile enterotoxin B subunit (LTB) bio-adjuvant, to develop an effective and reliable recombinant vaccine for NDV. In this study, the L(HN)2F protein, composed of the LTB bio-adjuvant and the immunogenic regions of the doubled Hemagglutinin Neuraminidase (HN-HN) and Fusion (F) epitope, was expressed in Escherichia coli. Subcutaneous injection was used to evaluate the humoral immune response in mice and the result was compared with B1 vaccine. The induction of strong humoral immune responses proved the strong immunoreactivity of the recombinant protein. The IgG elicited by the recombinant proteins was comparable to that of the commercial B1 vaccine against NDV, indicating its potential as a viable candidate for further development and evaluation as a recombinant vaccine against NDV.

A multi-source domain feature-decision dual fusion adversarial transfer network for cross-domain anti-noise mechanical fault diagnosis in sustainable city

Rotating machinery forms the critical backbone of infrastructure in a sustainable city, with bearings playing a pivotal role as key mechanical transmission components. Therefore, the health status of these bearings directly influences the safe operation of the infrastructure. Accurate and reliable diagnosis of defects in these components minimizes downtime, reduces maintenance costs, and prevents major accidents, ultimately providing insights in the construction and management of a sustainable city. Typically, in actual industrial scenarios, varying working conditions and various types of machines can result in significant discrepancies in the distribution of sample data. Moreover, the non-negligible noise may degrade the diagnostic performance. Therefore, realizing an accurate and reliable bearing diagnosis considering the cross-domain and noise environment remains a challenge. Leveraging the merits of information fusion and multi-source domain transfer learning, this article proposes a multi-source domain feature-decision dual fusion adversarial transfer network (DFATN) to break through the aforesaid limitations. Initially, an adversarial transfer framework is developed, incorporating novel feature matching evaluation and joint distribution difference losses. This framework is designed to facilitate the learning of feature invariants across domains and to enhance the sharing of domain-specific knowledge, even in noise. Relying on channel-spatial interactive feature fusion, a multi-scale feature extractor (MFE) is constructed to share the interaction and enhance the modeling of complex features in multiple dimensions. Additionally, a fault state-related decision fusion mechanism (SDF) is also implemented to integrate diagnostic information, significantly enhancing the generalization performance and robustness of the proposed network. By employing both public Paderborn University (PU) and laboratory-collected (Lab) datasets, the effectiveness and superiority of the proposed DFATN on bearing fault diagnosis are validated. For cross-working condition tasks, the proposed method realizes impressive performance, with average accuracies of 96.52% and 98.76% for Paderborn University (PU) and laboratory-collected (Lab) datasets, respectively. For cross-machine tasks, the average accuracy is 83.36%, outperforming other latest cross-domain fault diagnosis techniques.

Potent Inhibition and Rapid Photoactivation of Endogenous Bruton's Tyrosine Kinase Activity in Native Cells via Opto-Covalent Modulators.

Naturally, kinases exert their activities in a highly regulated fashion. A number of ingenious approaches have been developed to artificially control kinase activity by external stimuli, such as the incorporation of unnatural amino acids or the fusion of additional protein domains; however, methods that directly modulate endogenous kinases in native cells are lacking. Herein, we present a facile and potent method that takes advantage of recent developments in targeted covalent inhibitors and rapid light-mediated uncaging chemistry. Using an important drug target, Bruton's tyrosine kinase (BTK), as an example, these opto-covalent modulators successfully blocked the activity of endogenous BTK in native cells after simple incubation and washout steps. However, upon a few minutes of light irradiation, BTK activity was cleanly restored, and could be blocked again by conventional inhibitors. Promisingly, this photoactivation strategy easily worked in human peripheral blood mononuclear cells (hPBMCs).

Using Reporter Gene Assays to Screen and Identify Chemical Compounds that Modulate Estrogen Receptor Activity.

Estrogen receptor alpha (ERα) is a nuclear receptor that is expressed mainly in the breast, uterus, and ovary, among several other organs. ERα plays important roles in reproduction, mammary gland formation, and glucose homeostasis. Disruption of ERα may result in adverse outcomes, such as cancer, impaired fertility, and abnormal fetal growth. Therefore, identifying compounds that modulate ERα is of great interest due to their potential-endocrine disrupting capability and pharmaceutical applications. To rapidly test tens of thousands of compounds, high-throughput screening assays are essential. Here, we describe high-throughput screening methods, including plating and treatment of cells in 384-well and 1536-well plates and analysis of the resulting data. The two cell lines used, MCF7-VM7Luc4E2 and HEK293-ERα-bla, have been described previously. MCF7-VM7Luc4E2 cells are a stable luciferase reporter gene cell line expressing full-length endogenous estrogen receptor in the MCF7 cell line background, and HEK293-ERα-bla cells stably express an ERα ligand-binding domain/GAL4 DNA-binding domain fusion regulating a UAS β-lactamase reporter gene. These cell lines can be used to identify and confirm ERα modulators. Published 2024. This article is a U.S. Government work and is in the public domain in the USA. Current Protocols published by Wiley Periodicals LLC. Basic Protocol 1: Establishment of a high-throughput ERα reporter gene assay with luminescence readout to identify activators and inhibitors of estrogen receptor α Basic Protocol 2: Use of an orthogonal assay with fluorescence readout to confirm potential estrogen receptor activators or inhibitors.

Engineering a cleaved, prefusion-stabilized influenza B virus hemagglutinin by identification and locking of all six pH switches.

Vaccine components based on viral fusion proteins require high stability of the native prefusion conformation for optimal potency and manufacturability. In the case of influenza B virus hemagglutinin (HA), the stem's conformation relies on efficient cleavage. In this study, we identified six pH-sensitive regions distributed across the entire ectodomain where protonated histidines assume either a repulsive or an attractive role. Substitutions in these areas enhanced the protein's expression, quality, and stability in its prefusion trimeric state. Importantly, this stabilization enabled the production of a cleavable HA0, which is further processed into HA1 and HA2 by furin during exocytic pathway passage, thereby facilitating correct folding, increased stability, and screening for additional stabilizing substitutions in the core of the metastable fusion domain. Cryo-EM analysis at neutral and low pH revealed a previously unnoticed pH switch involving the C-terminal residues of the natively cleaved HA1. This switch keeps the fusion peptide in a clamped state at neutral pH, averting premature conformational shift. Our findings shed light on new strategies for possible improvements of recombinant or genetic-based influenza B vaccines.

Transfer Learning for Cross-City Traffic Prediction to Solve Data Scarcity

Deep learning models have demonstrated significant achievements in traffic prediction. However, their predictive performance substantially declines when faced with the scarcity of urban traffic data. Addressing the challenges of data scarcity and heterogeneity between cities, cross-city transfer learning has emerged as a promising solution. This paper proposes a domain adaptation cross-city model, which integrates traffic data with auxiliary urban data for domain adaptation in cross-city transfer learning. Specifically, we designed a domain fusion module to measure the differences between cities. Firstly, the knowledge extractor within the domain fusion module learns the knowledge from urban auxiliary data, such as road networks and points of interest, and calculates transferable knowledge. Then, dynamic time warping is used to measure the similarity of traffic time series. By combining these two aspects, we derive the domain differences between cities. Finally, the spatiotemporal network undergoes pre-learning using abundant data from the source city. According to the differences between cities, the model is fine-tuned to improve the adaptability of the model to the target domain. Experimental results on real-world data validate the effectiveness of the proposed model.

The Structural Diversity of Encapsulin Protein Shells.

Subcellular compartmentalization is a universal feature of all cells. Spatially distinct compartments, be they lipid- or protein-based, enable cells to optimize local reaction environments, store nutrients, and sequester toxic processes. Prokaryotes generally lack intracellular membrane systems and usually rely on protein-based compartments and organelles to regulate and optimize their metabolism. Encapsulins are one of the most diverse and widespread classes of prokaryotic protein compartments. They self-assemble into icosahedral protein shells and are able to specifically internalize dedicated cargo enzymes. This review discusses the structural diversity of encapsulin protein shells, focusing on shell assembly, symmetry, and dynamics. The properties and functions of pores found within encapsulin shells will also be discussed. In addition, fusion and insertion domains embedded within encapsulin shell protomers will be highlighted. Finally, future research directions for basic encapsulin biology, with a focus on the structural understand of encapsulins, are briefly outlined.

Exploring fusion domain: Advancing infrared and visible image fusion via IDFFN-GAN

Infrared (IR) and visible (VI) images are crucial in applications such as surveillance and night vision, where each modality provides complementary information—IR captures thermal details, while VI captures textures. Fusing these images is essential to combine their strengths, resulting in a more comprehensive and informative image. In this work, we introduce the “fusion domain” concept, a unique distribution that optimally blends IR and VI features. Our primary contribution is developing the Intermediate Domain Feature Fusion Network (IDFFN), which employs an MLP to learn the optimal domain factor for this fusion. Integrated with the IDFFN-GAN’s dual discriminators, our approach refines the fusion process to produce images that better preserve thermal and texture details from the proposed fusion and gradient domains. Experimental results demonstrate that our method achieves a 5% boost in Entropy (EN), a 50% rise in Spatial Frequency (SF), a 58% improvement in Average Gradient (AG), a 10% enhancement in Standard Deviation (SD), and a remarkable 198% increase in NAB/F compared to state-of-the-art techniques, ensuring superior retention of specific features from both modalities.

Fault Diagnosis of Low-Noise Amplifier Circuit Based on Fusion Domain Adaptation Method

The Low-Noise Amplifier (LNA) is a critical component of Radio Frequency (RF) receivers. Therefore, the accuracy of LNA fault diagnosis significantly impacts the overall performance of the entire RF receiver. Traditional LNA fault diagnosis is typically conducted under fixed conditions, but varying factors in practical applications often alter the circuit’s parameters and reduce diagnostic accuracy. To address the issue of decreased fault diagnosis accuracy under varying external or internal conditions, a fusion domain adaptation method based on Convolutional Neural Networks (CNNs), referred to as FDA, is proposed. Firstly, a domain-adaptive diagnostic model was established based on the feature extraction capabilities of CNNs. The powerful deep feature extraction capabilities of CNNs and the adaptability of domain adaptation methods to changing conditions are leveraged to enhance both the generalization ability of diagnostic models and the environmental adaptability of diagnostic techniques. Secondly, the fusion of feature-mapping domain adaptation and adversarial domain adaptation further enhances the convergence speed and diagnostic accuracy of the LNA cross-domain fault diagnosis model in the target domain. Finally, various cross-domain experiments were conducted. The FDA method achieved an average fault diagnosis rate of 90.19%, which represents an improvement of over 30% in accuracy compared to a CNN and also shows enhancements over individual domain-adaptation methods.

A Domain Generation Diagnosis Framework for Unseen Conditions Based on Adaptive Feature Fusion and Augmentation

Emerging deep learning-based fault diagnosis methods have advanced in the current industrial scenarios of various working conditions. However, the prerequisite of obtaining target data in advance limits the application of these models to practical engineering scenarios. To address the challenge of fault diagnosis under unseen working conditions, a domain generation framework for unseen conditions fault diagnosis is proposed, which consists of an Adaptive Feature Fusion Domain Generation Network (AFFN) and a Mix-up Augmentation Method (MAM) for both the data and domain spaces. AFFN is utilized to fuse domain-invariant and domain-specific representations to improve the model’s generalization performance. MAM enhances the model’s exploration ability for unseen domain boundaries. The diagnostic framework with AFFN and MAM can effectively learn more discriminative features from multiple source domains to perform different generalization tasks for unseen working loads and machines. The feasibility of the proposed unseen conditions diagnostic framework is validated on the SDUST and PU datasets and achieved peak diagnostic accuracies of 94.15% and 93.27%, respectively.

Development of adeno-associated viral vectors targeting cardiac fibroblasts for efficient in vivo cardiac reprogramming

Overexpression of cardiac reprogramming factors, including GATA4, HAND2, TBX5, and MEF2C (GHT/M), can directly reprogram cardiac fibroblasts (CFs) into induced cardiomyocytes (iCMs). Adeno-associated virus (AAV) vectors are widely used clinically, and vectors targeting cardiomyocytes (CMs) have been extensively studied. However, safe and efficient AAV vectors targeting CFs for invivo cardiac reprogramming remain elusive. Therefore, we screened multiple AAV capsids and promoters to develop efficient and safe CF-targeting AAV vectors for invivo cardiac reprogramming. AAV-DJ capsids containing periostin promoter (AAV-DJ-Postn) strongly and specifically expressed transgenes in resident CFs in mice after myocardial infarction (MI). Lineage tracing revealed that AAV-DJ-Postn vectors expressing GHT/M reprogrammed CFs into iCMs, which was further increased 2-fold using activated MEF2C via the fusion of the powerful MYOD transactivation domain (M-TAD) with GHT (AAV-DJ-Postn-GHT/M-TAD). AAV-DJ-Postn-GHT/M-TAD injection improved cardiac function and reduced fibrosis after MI. Overall, we developed new AAV vectors that target CFs for cardiac reprogramming.

Fault diagnosis of photovoltaic array with multi-module fusion under hyperparameter optimization

Photovoltaic (PV) arrays’ random and intermittent output characteristics impact power system safety. To improve the performance of the PV array fault diagnosis model, a novel online fault monitoring technique is introduced. (1) Fault diagnostic model construction: Significant differences in PV arrays’ I-V and P-V curves under various fault conditions led to constructing a 3D channel feature map based on I, V, and P features. (2) Multi-source information fusion network (MSIFN): this multi-module fusion model includes a time–frequency domain fusion module (TDFM), a multi-feature shuffle expansion convolution module (MSECM), a parameter-free parallel hybrid attention enhancement module, and a multi-scale mixed pooling fusion classification module (MMPCM). (3) Multi-strategy fusion whale optimization algorithm (MSFWOA): addressing the original WOA’s deficiencies, we designed time control, parameter modification, and greedy control strategies based on lens imaging to optimize MSIFN’s hyper-parameters. Experimental results show that the MSFWOA-MSIFN model excels in PV array fault diagnosis (Paccuracy=Pprecision=Precall = 99.92 %). In three types of noise experiments with 15 dB, 25 dB, and 30 dB, the average performance index remained above 99 %. In practical experiments, the average performance indices werePaccuracy = 97.53 %, Pprecision = 97.32 %, andPrecall = 97.41 %, further demonstrating its excellent diagnostic performance. This model effectively diagnoses various faults in PV arrays, providing scientific and theoretical support for PV system operations.

Efficient and adaptive secure cross-domain recommendations

In recommendation systems, cross-domain recommendation (CDR) has emerged as an effective method for solving the problem of data sparsity and cold start. However, challenges such as data heterogeneity, data security, and recommendation accuracy are faced by existing CDR methods. An efficient and adaptive secure cross-domain recommendation (EAAS-CDR) model is proposed. First, the Automatic Encoder–Decoder (AutoED) model is proposed to address variations in user–item interaction quantity and item characteristics across various domains. The AutoED model enables adaptive and secure cross-domain recommendations by modeling user preferences across various domains, unifying vector dimensions, and avoiding the release of the rating matrix. An efficient cross-domain privacy protection method is proposed to resist data leakage. To ensure feature invariance for knowledge transfer, the inherent low perturbation characteristics of the Johnson–Lindenstrauss Transform mechanism are leveraged and combined with the output of our encoder. Finally, domain relevance measurement metrics and a domain fusion algorithm are proposed to use the source domain knowledge in the target domain and improve the recommendation accuracy of the model. Experimental results demonstrate that our approach is not only more efficient but also more secure than existing CDR approaches.

GNN-based multi-source domain prototype representation for cross-subject EEG emotion recognition

Emotion recognition based on electroencephalography (EEG) signals is a major area of affective computing. However, the existence of distributional differences between subjects has greatly hindered the large-scale application of EEG emotion recognition techniques. Most of the existing cross-subject methods primarily concentrate on treating multiple subjects as a single source domain. These methods lead to significant distributional differences within the source domain, which hinder the model’s ability to generalise effectively to target subjects. In this paper, we propose a new method that combines graph neural network-based prototype representation of multiple source domains with clustering similarity loss. It consists of three parts: multi-source domain prototype representation, graph neural network and loss. Multi-source domain prototype representation treats different subjects in the source domain as sub-source domains and extracts prototype features, which learns a more fine-grained feature representation. Graph neural network can better model the association properties between prototypes and samples. In addition, we propose a similarity loss based on clustering idea. The loss makes maximum use of similarity between samples in the target domain while ensuring that the classification performance does not degrade. We conduct extensive experiments on two benchmark datasets, SEED and SEED IV. The experimental results validate the effectiveness of the proposed multi-source domain fusion approach and indicate its superiority over existing methods in cross-subject classification tasks.

Acoustically shaped DNA-programmable materials

Recent advances in DNA nanotechnology allow for the assembly of nanocomponents with nanoscale precision, leading to the emergence of DNA-based material fabrication approaches. Yet, transferring these nano- and micron-scale structural arrangements to the macroscale morphologies remains a challenge, which limits the development of materials and devices based on DNA nanotechnology. Here, we demonstrate a materials fabrication approach that combines DNA-programmable assembly with actively driven processes controlled by acoustic fields. This combination provides a prescribed nanoscale order, as dictated by equilibrium assembly through DNA-encoded interactions, and field-shaped macroscale morphology, as regulated by out-of-equilibrium materials formation through specific acoustic stimulation. Using optical and electron microscopy imaging and x-ray scattering, we further revealed the nucleation processes, domain fusion, and crystal growth under different acoustically stimulated conditions. The developed approach provides a pathway for the fabrication of complexly shaped macroscale morphologies for DNA-programmable nanomaterials by controlling spatiotemporal characteristics of the acoustic fields.

Phase I studies of davoceticept (ALPN-202), a PD-L1-dependent CD28 co-stimulator and dual PD-L1/CTLA-4 inhibitor, as monotherapy and in combination with pembrolizumab in advanced solid tumors (NEON-1 and NEON-2)

BackgroundDavoceticept (ALPN-202) is an Fc fusion of a CD80 variant immunoglobulin domain designed to mediate programmed death-ligand 1 (PD-L1)-dependent CD28 co-stimulation while inhibiting the PD-L1 and cytotoxic T-lymphocyte-associated antigen 4...

Positive residues of the SARS-CoV-2 fusion domain are key contributors to the initiation of membrane fusion

SARS-CoV-2 is one of the most infectious viruses ever recorded. Despite a plethora of research over the last several years, the viral lifecycle is still not well understood, particularly membrane fusion. This process is initiated by the fusion domain (FD), a highly conserved stretch of amino acids consisting of a fusion peptide (FP) and fusion loop (FL), which in synergy perturbs the target cells lipid membrane to lower the energetic cost necessary for fusion. In this study, through a mutagenesis-based approach, we have investigated the basic residues within the FD (K825, K835, R847, K854) utilizing an in vitro fusion assay and 19F NMR, validated by traditional 13C 15N techniques. Alanine and charge conserving mutants revealed every basic residue plays a highly specific role within the mechanism of initiating fusion. Intriguingly, K825A led to increased fusogenecity which was found to be correlated to the number of amino acids within helix one, further implicating the role of this specific helix within the FD’s fusion mechanism. This work has found basic residues to be important within the FDs fusion mechanism and highlights K825A, a specific mutation made within the FD of the SARS-CoV-2 spike protein, as requiring further investigation due to its potential to contribute to a more virulent strain of SARS-CoV-2.