Multiple Recognition Research Articles

BRCA1-BARD1 combines multiple chromatin recognition modules to bridge nascent nucleosomes.

Chromatin association of the BRCA1-BARD1 heterodimer is critical to promote homologous recombination repair of DNA double-strand breaks (DSBs) in S/G2. How the BRCA1-BARD1 complex interacts with chromatin that contains both damage induced histone H2A ubiquitin and inhibitory H4K20 methylation is not fully understood. We characterised BRCA1-BARD1 binding and enzymatic activity to an array of mono- and di-nucleosome substrates using biochemical, structuraland single molecule imaging approaches. We found that the BRCA1-BARD1 complex preferentially interacts and modifies di-nucleosomes over mono-nucleosomes, allowing integration of H2A Lys-15 ubiquitylation signals with other chromatin modifications and features. Using high speed- atomic force microscopy (HS-AFM) to monitor how the BRCA1-BARD1 complex recognises chromatin in real time, we saw a highly dynamic complex that bridges two nucleosomes and associates with the DNA linker region. Bridging is aided by multivalent cross-nucleosome interactions that enhance BRCA1-BARD1 E3 ubiquitin ligase catalytic activity. Multivalent interactions across nucleosomes explain how BRCA1-BARD1 can recognise chromatin that retains partial di-methylation at H4 Lys-20 (H4K20me2), a parental histone mark that blocks BRCA1-BARD1 interaction with nucleosomes, to promote its enzymatic and DNA repair activities.

MLPs Are All You Need for Human Activity Recognition

Convolution, recurrent, and attention-based deep learning techniques have produced the most recent state-of-the-art results in multiple sensor-based human activity recognition (HAR) datasets. However, these techniques have high computing costs, restricting their use in low-powered devices. Different methods have been employed to increase the efficiency of these techniques; however, this often results in worse performance. Recently, pure multi-layer perceptron (MLP) architectures have demonstrated competitive performance in vision-based tasks with lower computation costs than other deep-learning techniques. The MLP-Mixer is a pioneering pureMLP architecture that produces competitive results with state-of-the-art models in computer vision tasks. This paper shows the viability of the MLP-Mixer in sensor-based HAR. Furthermore, experiments are performed to gain insight into the Mixer modules essential for HAR, and a visual analysis of the Mixer’s weights is provided, validating the Mixer’s learning capabilities. As a result, the Mixer achieves F1 scores of 97%, 84.2%, 91.2%, and 90% on the PAMAP2, Daphnet Gait, Opportunity Gestures, and Opportunity Locomotion datasets, respectively, outperforming state-of-the-art models in all datasets except Opportunity Gestures.

Multiple paddy disease recognition methods based on deformable transformer attention mechanism in complex scenarios

Paddy disease recognition presents challenges in the agricultural industry, and existing algorithms struggle to accurately identify diseases in complex scenarios. In this paper, we propose a precise object detection framework to address the challenges of severe overlap, multi-disease detection, morphological irregularities, multi-scale object classification, and complex scenarios in real-world environments in paddy disease detection. The proposed model is based on an improved version of the DEtection TRansformer (Detr) algorithm. The enhanced network architecture fuses multi-scale features by adding a feature fusion module after the backbone network, which is able to retain more original information of the images and greatly improves the detection accuracy; the use of deformable attention module greatly reduces the computational complexity of the model. To evaluate the PDN, a dedicated paddy disease detection dataset with 1200 images is created. Experimental results demonstrate that the proposed model obtained a precision value of 100%, a recall value of 89.3%, F1-score of 94.3%, and a mean average precision (mAP) value of 60.2%. The model outperforms the existing state-of-the-art detection models in detection accuracy.

Application of multiple difference feature network and speech recognition in dance training system

Application of multiple difference feature network and speech recognition in dance training system

Future Pharmacological Prospectives and Multiple Diagnostic Recognition of Traditional Medicinal Plant Ficus racemosa

A significant medicinal plant called ‘Ficus racemosa’ of the Moraceae species can be found in Southeast Asia, Australia, and India, Mainly in the states of Maharashtra, Gujarat, Uttar Pradesh, Karnataka, and Tamil Nadu. It is frequently referred to as “gular.” Due to the presence of bergenin as a flavonoid, it acts as an anticancer. Numerous isolated active components from various portions of this plant have been discovered to possess advantageous pharmacological characteristics. According to a literature review, it has various pharmacological actions, including liver-protective activity, anti-HIV, antidiabetic, antidiarrheal, antioxidant, antipyretic, anti-inflammatory, antifungal, anticancer and antibacterial activities. This review study does a good job of discussing this particular plant’s traditional usage, phytochemistry, pharmacology, and toxicity.

A surface multiple imprinting layers membrane with well-oriented recognition sites for selective separation of chlorogenic acid from Ficus carica L.

Chlorogenic acid (CGA), known as an important natural antioxidative compound in Ficus carica L, has valuable application prospects on health food, functional food, nutrition and dietary formulations. In this study, a surface multiple imprinting layers membrane (SMILM) was developed and applied to separate CGA from F. carica. Two different imprinting layers were integrated onto the membrane surface in sequence. The first imprinting layer was formed by dopamine polymerization and the second imprinting layer was fabricated by atom transfer radical polymerization (ATRP) and boronate affinity. The prepared SMILM with well-oriented multiple recognition sites exhibited high adsorption capacity (52.08 mg·g−1 in 60 min) for CGA and specific selectivity with imprinting factor (IF) of 3.06. Furthermore, the recognition mechanism of SMILM was clarified by molecular simulation and NMR. The SMILM was successfully applied to separate CGA from the fruits, peels and leaves of F. carica with recoveries of 90.22, 83.31 and 84.95 %, respectively.

Research on Multi-Source Simultaneous Recognition Technology Based on Sagnac Fiber Optic Sound Sensing System

To solve the problem of multiple sound recognition in the application of Sagnac optical fiber acoustic sensing system, a multi-source synchronous recognition algorithm was proposed, which combined the VMD (variational modal decomposition) algorithm and MFCC (Mel-frequency cepstral coefficient algorithm) algorithm to pre-process the photoacoustic sensing signal, and uses BP neural network to recognize the photoacoustic sensing signal. The modal analysis and feature extraction theory of photoacoustic sensing signal based on the VMD and MFCC algorithms were presented. The signal recognition theory analysis and system recognition program design were completed based on the BP neural network. Signal acquisition of different sounds and verification experiments of the recognition system have been carried out in a laboratory environment based on the Sagnac fiber optic sound sensing system. The experimental results show that the proposed optical fiber acoustic sensing signal recognition algorithm has a simultaneous recognition rate better than 96.5% for six types of sounds, and the optical acoustic signal recognition takes less than 5.3 s, which has the capability of real-time sound detection and recognition, and provides the possibility of further application of the Sagnac-based optical fiber acoustic sensing system.

Biosynthesized gold nanoparticles that activate Toll-like receptors and elicit localized light-converting hyperthermia for pleiotropic tumor immunoregulation

Manipulating the tumor immune contexture towards a more active state can result in better therapeutic outcomes. Here we describe an easily accessible bacterial biomineralization-generated immunomodulator, which we name Ausome (Au + [exo]some). Ausome comprises a gold nanoparticle core covered by bacterial components; the former affords an inducible hyperthermia effect, while the latter mobilizes diverse immune responses. Multiple pattern recognition receptors actively participate in Ausome-initiated immune responses, which lead to the release of a broad spectrum of pro-inflammatory cytokines and the activation of effector immune cells. Upon laser irradiation, tumor-accumulated Ausome elicits a hyperthermic response, which improves tissue blood perfusion and contributes to enhanced infiltration of immunostimulatory modules, including cytokines and effector lymphocytes. This immune-modulating strategy mediated by Ausome ultimately brings about a comprehensive immune reaction and selectively amplifies the effects of local antitumor immunity, enhancing the efficacy of well-established chemo- or immuno-therapies in preclinical cancer models in female mice.

Relative-position embedding based spatially and temporally decoupled Transformer for action recognition

Recognition of human actions is to classify actions in a video. Recently, Vision Transformer (ViT) has been applied to action recognition. However, the Vision Transformer is unsuitable for high-resolution input videos due to the constraint of computing power since ViT splits frames into fixed-size patches embedded (i.e., tokens) with absolute-position information and adopts a pure Transformer encoder to model the relationships among these tokens. To address this issue, we propose a relative-position embedding based spatially and temporally decoupled Transformer (RPE-STDT) for action recognition, which can capture spatial–temporal information by stacked self-attention layers. The proposed RPE-STDT model consists of two separate series of Transformer encoders. The first series of encoders is the spatial Transformer encoders, which model interactions between tokens extracted from the same temporal index. The second series of encoders is the temporal Transformer encoders, which model interactions across time dimensions with a subsampling strategy. Furthermore, we replace the absolute-position embeddings in the Vision Transformer encoders with the proposed relative-position embeddings to capture the order of the embedded tokens to reduce computational costs. Finally, we conduct thorough ablation studies. Our RPE-STDT achieves state-of-the-art results on multiple action recognition datasets, exceeding prior convolution and Transformer-based networks.

Exploration of Deep Learning Models for Video Based Multiple Human Activity Recognition

Human Activity Recognition (HAR) with Deep Learning is a challenging and a highly demanding classification task. Complexity of the activity detection and the number of subjects are the main issues. Data mining approaches improved decision-making performance. This work presents one such model for Human activity recognition for multiple subjects carrying out multiple activities. Involving real time datasets, the work developed a rapid algorithm for minimizing the problems of neural networks classifier. An optimal feature extraction happens and develops a multi-modal classification technique and predicts solutions with better accuracy when compared to other traditional methods. This paper discussing on HAR prediction in four phases namely (i) Depthwise Separable Convolution with BiLSTM (DSC-BLSTM); (ii) Enhanced Bidirectional Grated Recurrent Unit with Long Short Term Memory (BGRU-LSTM); (iii) Enhanced TimeSformer Model with Multi-Layer Perceptron Neural Networks classification and (iv) Filtering Single Activity Recognition are described.In comparison to previous efforts like the DSC-BLSTM and BGRU-LSTM classifications, the experimental result of the ETMLP classification attained 98.90%, which was more efficient. The end outcome revealed that the new model performed better in terms of accuracy than the other models.

“OR” logic gate multiplexed photoelectrochemical sensor for high-risk human papillomaviruses: “One pot” recombinase polymerase amplification and logic discrimination

Multiple targets analysis in complex samples is of great importance in medical and health sciences. Limited by independent laborious operational procedures, multiple targets determination remains a challenge. Herein, we report an “OR” logic gate multiplexed photoelectrochemical (PEC) sensor based on “one pot” recombinase polymerase amplification (RPA) strategy. “One pot” RPA triggers exponential growth of multiple DNA in complex samples. Subsequently, the amplification products interact separately with lambda exonuclease (λ exo) or Cas12a-crRNA. Following the multiple targets recognition event, the dual enzyme-mediated cleavage separates the signal labels from the photocathode. The resulting photocurrent change is utilized for logical discrimination and detection. The feasibility of the sensor is verified by analyzing the two typical duplex DNA (high-risk human papillomaviruses (HPV)). Ultralow detection limit (0.088 fg/μL, 0.081 fg/μL) with broad detection range (0.1 fg/μL to 10 ng/μL, 0.1 fg/μL to 1 ng/μL) for HPV16 and HPV18 are obtained. Eliminating instrumentation constraints (light source/potential modulation) and simplifying operation procedures, this work opens an avenue for developing multiplexed sensing devices for clinical diagnosis and disease treatment.

Unravelling dispersion forces in liquid-phase enantioseparation. Part I: Impact of ferrocenyl versus phenyl groups

BackgroundHighly ordered chiral secondary structures as well as multiple (tunable) recognition sites are the keys to success of polysaccharide carbamate-based chiral selectors in enantioseparation science. Hydrogen bonds (HBs), dipole-dipole, and π-π interactions are classically considered the most frequent noncovalent interactions underlying enantioselective recognition with these chiral selectors. Very recently, halogen, chalcogen and π-hole bonds were also identified as interactions working in polysaccharide carbamate-based selectors to promote enantiomer distinction. On the contrary, the function of dispersion interactions in this field was not explored so far. ResultsThe enantioseparation of chiral ferrocenes featuring chiral axis or chiral plane as stereogenic elements was performed by comparing five polysaccharide carbamate-based chiral columns, with the aim to identify enantioseparation outcomes that could be reasonably determined by dispersion forces, making available a reliable experimental data set for future theoretical studies to confirm the heuristic hypothesis. The effects of mobile phase polarity and temperature on the enantioseparation were considered, and potential recognition sites on analytes and selectors were evaluated by electrostatic potential (V) analysis and molecular dynamics (MD). In this first part, the enantioseparation of 3,3′-dibromo-5,5′-bis-ferrocenylethynyl-4,4′-bipyridine bearing two ferrocenylethynyl units linked to an axially chiral core was performed and compared to that of the analyte featuring the same structural motif with two phenyl groups in place of the ferrocenyl moieties. The results of this study showed the superiority of the ferrocenyl compared to the phenyl group, as a structural element favouring enantiodifferentiation. Significance and noveltyEven if dispersion (London) forces have been envisaged acting in liquid-phase enantioseparations, focused studies to explore possible contributions of dispersion forces with polysaccharide carbamate-based selectors are practically missing. This study allowed us to collect experimental information that support the involvement of dispersion forces as contributors to liquid-phase enantioseparation, paving the way to a new picture in this field.

Effector-dependent structural transformation of a crystalline framework with allosteric effects on molecular recognition ability

Structurally flexible porous crystals that combine high regularity and stimuli responsiveness have received attracted attention in connection with natural allostery found in regulatory systems of activity and function in biological systems. Porous crystals with molecular recognition sites in the inner pores are particularly promising for achieving elaborate functional control, where the local binding of effectors triggers their distortion to propagate throughout the structure. Here we report that the structure of a porous molecular crystal can be allosterically controlled by local adsorption of effectors within low-symmetry nanochannels with multiple molecular recognition sites. The exchange of effectors at the allosteric site triggers diverse conversion of the framework structure in an effector-dependent manner. In conjunction with the structural conversion, it is also possible to switch the molecular affinity at different recognition sites. These results may provide a guideline for the development of supramolecular materials with flexible and highly-ordered three-dimensional structures for biological applications.

Direct training high-performance spiking neural networks for object recognition and detection.

The spiking neural network (SNN) is a bionic model that is energy-efficient when implemented on neuromorphic hardwares. The non-differentiability of the spiking signals and the complicated neural dynamics make direct training of high-performance SNNs a great challenge. There are numerous crucial issues to explore for the deployment of direct training SNNs, such as gradient vanishing and explosion, spiking signal decoding, and applications in upstream tasks. To address gradient vanishing, we introduce a binary selection gate into the basic residual block and propose spiking gate (SG) ResNet to implement residual learning in SNNs. We propose two appropriate representations of the gate signal and verify that SG ResNet can overcome gradient vanishing or explosion by analyzing the gradient backpropagation. For the spiking signal decoding, a better decoding scheme than rate coding is achieved by our attention spike decoder (ASD), which dynamically assigns weights to spiking signals along the temporal, channel, and spatial dimensions. The SG ResNet and ASD modules are evaluated on multiple object recognition datasets, including the static ImageNet, CIFAR-100, CIFAR-10, and neuromorphic DVS-CIFAR10 datasets. Superior accuracy is demonstrated with a tiny simulation time step of four, specifically 94.52% top-1 accuracy on CIFAR-10 and 75.64% top-1 accuracy on CIFAR-100. Spiking RetinaNet is proposed using SG ResNet as the backbone and ASD module for information decoding as the first direct-training hybrid SNN-ANN detector for RGB images. Spiking RetinaNet with a SG ResNet34 backbone achieves an mAP of 0.296 on the object detection dataset MSCOCO.

BIK1 protein homeostasis is maintained by the interplay of different ubiquitin ligases in immune signaling

Pathogen-associated molecular patterns (PAMPs) trigger plant innate immunity that acts as the first line of inducible defense against pathogen infection. A receptor-like cytoplasmic kinase BOTRYTIS-INDUCED KINASE 1 (BIK1) functions as a signaling hub immediately downstream of multiple pattern recognition receptors (PRRs). It is known that PLANT U-BOX PROTEIN 25 (PUB25) and PUB26 ubiquitinate BIK1 and mediate BIK1 degradation. However, how BIK1 homeostasis is maintained is not fully understood. Here, we show that two closely related ubiquitin ligases, RING DOMAIN LIGASE 1 (RGLG1) and RGLG2, preferentially associate with the hypo-phosphorylated BIK1 and promote the association of BIK1 with the co-receptor for several PRRs, BRI1-ASSOCIATED RECEPTOR KINASE1 (BAK1). PUB25 interacts with RGLG2 and mediates its degradation. In turn, RGLG2 represses the ubiquitin ligase activity of PUB25. RGLG1/2 suppress PUB25-mediated BIK1 degradation, promote BIK1 protein accumulation, and positively regulate immune signaling in a ubiquitin ligase activity-dependent manner. Our work reveals how BIK1 homeostasis is maintained by the interplay of different ubiquitin ligases.

Self-powered dual-mode sensing strategy based on graphdiyne and DNA nanoring for sensitive detection of tumor biomarker

MicroRNA-21 (miRNA-21) is currently the only known oncogenic miRNA that is upregulated in almost all malignant tumors and exhibits a broad spectrum of tumor recognition characteristics. It holds significant value in the early diagnosis, malignant degree assessment, and prognostic evaluation of tumors. In this study, a novel dual-mode self-powered sensing platform is developed using Au nanoparticles/graphdiyne as the electrode substrate and combined with DNA nanoring for highly sensitive and specific detection of miRNA-21. The DNA nanoring structure, which is easy to prepare and contains multiple recognition sites, induces significant electrochemical/colorimetric signal responses of the signaling molecule methylene blue. Under optimal conditions, the linear ranges of the electrochemical and colorimetric detection modes of this self-powered sensor are 0.1 fM-100 pM and 0.1 fM-10 nM, respectively, with the detection limits of 35.1 aM and 61.6 aM (S/N=3). This strategy provides a new reference for the sensitive detection of microRNA and has immense potential for application in the screening and detection of clinical nucleic acid diseases.

NeuroCNN_GNB: an ensemble model to predict neuropeptides based on a convolution neural network and Gaussian naive Bayes.

Neuropeptides contain more chemical information than other classical neurotransmitters and have multiple receptor recognition sites. These characteristics allow neuropeptides to have a correspondingly higher selectivity for nerve receptors and fewer side effects. Traditional experimental methods, such as mass spectrometry and liquid chromatography technology, still need the support of a complete neuropeptide precursor database and the basic characteristics of neuropeptides. Incomplete neuropeptide precursor and information databases will lead to false-positives or reduce the sensitivity of recognition. In recent years, studies have proven that machine learning methods can rapidly and effectively predict neuropeptides. In this work, we have made a systematic attempt to create an ensemble tool based on four convolution neural network models. These baseline models were separately trained on one-hot encoding, AAIndex, G-gap dipeptide encoding and word2vec and integrated using Gaussian Naive Bayes (NB) to construct our predictor designated NeuroCNN_GNB. Both 5-fold cross-validation tests using benchmark datasets and independent tests showed that NeuroCNN_GNB outperformed other state-of-the-art methods. Furthermore, this novel framework provides essential interpretations that aid the understanding of model success by leveraging the powerful Shapley Additive exPlanation (SHAP) algorithm, thereby highlighting the most important features relevant for predicting neuropeptides.

Immunoinformatics and MD-simulation data suggest that Omicron spike epitopes are more interacting to IgG via better MHC recognition than Delta variant

BackgroundRecently, in Nov 2021, in South Africa, the SARS CoV-2 variant Omicron was found to be highly infectious and transmissible but with the least fatality. It occupies the nasopharynx-oropharynx and easily spreads. The epidemiological data/reports suggest that several vaccines failed to neutralize Omicron. It has a large number of spike mutations and the RNA/protein vaccines were developed from its predecessors that may justify its escape in most neutralization reactions. Its lower immuno-suppression/cytokine-storming/inflammatory-response effects need exploration. ObjectivesIn the current study, we attempted to delineate the comparative interaction of different variants’ spikes with multiple recognition sites on IgG and HLA-typing of MHC class and I and II. MethodsAll SARS-CoV-2 spike-proteins/human-IgG/MHC-I & II were obtained from the NCBI/ PDB/GISAID database. Initial 3D-structures of the unavailable proteins were constructed by Homology-Modeling (Swissmodel-Expasy) and optimized (PROCHECK). Molecular-docking of spike-IgG/spike- I & MHC-II was performed (HADDOCK2.4/HawkDock) with active-residue screening (CPORT). Antigenicity of epitopes was determined (Vaxigen v2.0-server) and the epitope-model prepared (PEP-FOLD3-server). The binding-affinity/biological-interfaces/visualize were performed (PRODIGY-PyMOL2). We also examined the genesis of feasible transition pathways of functional docked complexes (iMODs) of MHC with different epitopes and antibodies of IgG with different variants. Further, Molecular-Dynamic-Simulation was performed by GROMACS 2023.1 software package. The MD-simulation was run with 100 ns (300 k-heating/1-atm pressure). ResultsSurface-area with interactomes, H-bonding and polar/non-polar bonding were the highest in Omicron spike-IgG interaction. Unlike other variants, both the L and H chains of at least three different recognition sites of IgG interact with the N-terminal and C-terminal RBD of the S1-portion and partially bind to S2. In other cases, binding was observed in either NTD or CTD with a lesser number of bonding-interactomes, especially in Delta spike-Ab interaction. In the case of MHC class-I & II, the highest binding affinity/surface was noticed by Omicron and least by the Delta variant. The MD simulation data of lower RMSD values of the Delta and Omicron variants indicate improved structural stability and less departure from the initial conformation. Better binding to the IgG and MHC molecules explains Omicron’s little ability in immune invasion.

Advancements in Gas Recognition Techniques for Electronic Nose Systems: A Comparative Review of Classical Methods and Spiking Neural Networks

An electronic nose (E-nose) system's ability to recognize multivariate responses from gas sensors in a variety of applications necessitates gas recognition. Principal component analysis (PCA) and other traditional gas recognition methods have been widely used in E-nose systems for decades. ANNs have transformed the field of E- nose, particularly spiking neural networks (SNNs), significantly in recent years. In this paper, we compare and contrast recent E-nose gas recognition techniques in terms of algorithms and hardware implementations. Each classical gas recognition method has a relatively fixed framework and few parameters, making it easy to design. It works well with few gas samples but poorly with multiple gas recognition when noise is present. Keywords: Gas detection, electronic nose, artificial neural network, and spiking neural network.

The translational repressor Glorund uses interchangeable RNA recognition domains to recognize Drosophila nanos.

The Drosophila melanogaster protein Glorund (Glo) represses nanos (nos) translation and uses its quasi-RNA recognition motifs (qRRMs) to recognize both G-tract and structuredUA-rich motifs within the nos translational control element (TCE). We showed previously that each of the three qRRMs is multifunctional, capable of binding to G-tract and UA-rich motifs, yet if and how the qRRMs combine to recognize the nos TCE remained unclear. Here we determined solution structures of a nos TCEI_III RNA containing the G-tract and UA-rich motifs. The RNA structure demonstrated that a single qRRM is physically incapable of recognizing both RNA elements simultaneously. In vivo experiments further indicated that any two qRRMs are sufficient to repress nos translation. We probed interactions of Glo qRRMs with TCEI_III RNA using NMR paramagnetic relaxation experiments. Our in vitro and in vivo data support a model whereby tandem Glo qRRMs are indeed multifunctional and interchangeable for recognition of TCE G-tract or UA-rich motifs. This study illustrates how multiple RNA recognition modules within an RNA-binding protein may combine to diversify the RNAs that are recognized and regulated.