Shared Features Research Articles

Comparing alpha-synuclein-interactomes between multiple systems atrophy and Parkinson's disease reveals unique and shared pathological features.

Primary synucleinopathies, such as Parkinson's disease (PD), Dementia with Lewy bodies (DLB), and multiple system atrophy (MSA), are neurodegenerative disorders with some shared clinical and pathological features. Aggregates of alpha-synuclein (αsyn) phosphorylated at serine 129 (PSER129) are the hallmark of synucleinopathies, which for PD/DLB are found predominantly in neurons (Neuronal cytoplasmic inclusions "NCIs"), but for MSA, aggregates are primarily found in oligodendroglia (Glial cytoplasmic inclusions "GCIs"). It remains unclear if the distinct pathological presentation of PD/DLB and MSA are manifestations of distinct or shared pathological processes. We hypothesize that the distinct synucleinopathies MSA and PD/DLB share common molecular features. Using the in-situ proximity labeling technique biotinylation by antibody recognition (BAR), we compare aggregated αsyn-interactomes (BAR-PSER129) and total αsyn-interactomes (BAR-MJFR1) between MSA (n=5) and PD/DLB (n=10) in forebrain and midbrain structures. For BAR-PSER129 and BAR-MJFR1 captures, αsyn was the most significantly enriched protein in PD/DLB and MSA. In PD/DLB, BAR-PSER129 identified 194 αsyn-aggregate-interacting proteins, while BAR-MJFR1 identified 245 αsyn interacting proteins. In contrast, in the MSA brain, only 38 and 175 proteins were identified for each capture, respectively. When comparing MSA and PD/DLB, a high overlap (59.5%) was observed between BAR-MJFR1 captured proteins, whereas less overlap (14.4%) was observed for BAR-PSER129. Direct comparison between MSA and PD/DLB revealed 79 PD/DLB-associated proteins and only three MSA-associated proteins (CBR1, CRYAB, and GFAP). Pathway enrichment analysis revealed PD/DLB interactions were dominated by vesicle/SNARE-associated pathways, in contrast to MSA, which strongly enriched for metabolic/catabolic, iron, and cellular oxidant detoxification pathways. A subnetwork of cytosolic antioxidant enzymes called peroxiredoxins drove cellular detoxification pathways in MSA. A common network of 26 proteins, including neuronal-specific proteins (e.g., SNYGR3) with HSPA8 at the core, was shared between MSA and DLB/PD. Extracellular exosome pathways were universally enriched regardless of disease or BAR target protein. Synucleinopathies have divergent and convergent αsyn-aggregate interactions, indicating unique and shared pathogenic mechanisms. MSA uniquely involves oxidant detoxification processes in glial cells, while vesicular processes in neurons dominate PD/DLB. Shared interactions, specifically SNYGR3 (i.e., a neuronal protein), between MSA and PD/DLB suggest neuronal axons origin for both diseases. In conclusion, we provide αsyn aggregates protein interaction maps for two distinct synucleinopathies.

A Multimodal Sentiment Analysis Model for Graphic Texts Based on Deep Feature Interaction Networks

Due to the widespread adoption of social networks, image-text comments have become a prevalent mode of emotional expression compared to traditional text descriptions. However, there are currently two major challenges. The first is the question of how to extract rich representations effectively from both text and images, and the second is the question of how to extract cross-modal shared emotion features. This study proposes a multimodal sentiment analysis method based on a deep feature interaction network (DFINet). It leverages word-to-word graphs and deep attention interaction networks (DAIN) to learn text representations effectively from multiple subspaces. Additionally, it introduces a cross-modal attention interaction network to extract cross-modal shared emotion features efficiently. This approach helps alleviate the difficulties associated with acquiring image-text features and representing cross-modal shared emotion features. Experimental results on the Yelp dataset demonstrate the effectiveness of the DFINet method.

Shared neutrophil and T cell dysfunction is accompanied by a distinct interferon signature during severe febrile illnesses in children

Severe febrile illnesses in children encompass life-threatening organ dysfunction caused by diverse pathogens and other severe inflammatory syndromes. A comparative approach to these illnesses may identify shared and distinct features of host immune dysfunction amenable to immunomodulation. Here, using immunophenotyping with mass cytometry and cell stimulation experiments, we illustrate trajectories of immune dysfunction in 74 children with multi-system inflammatory syndrome in children (MIS-C) associated with SARS-CoV-2, 30 with bacterial infection, 16 with viral infection, 8 with Kawasaki disease, and 42 controls. We explore these findings in a secondary cohort of 500 children with these illnesses and 134 controls. We show that neutrophil activation and apoptosis are prominent in multi-system inflammatory syndrome, and that this is partially shared with bacterial infection. We show that memory T cells from patients with multi-system inflammatory syndrome and bacterial infection are exhausted. In contrast, we show viral infection to be characterized by a distinct signature of decreased interferon signaling and lower interferon receptor gene expression. Improved understanding of immune dysfunction may improve approaches to immunomodulator therapy in severe febrile illnesses in children.

The overlapping of phenotypes in Wiedemann-Steiner, Kleefstra and Coffin-Siris syndromes: a study of eleven patients

BackgroundSome chromatinopathies may present with common clinical findings (intellectual disability, brain and limb malformation, facial dysmorphism). Furthermore, one of their cardinal shared features is growth dysregulation.We aimed to assess and deepen this resemblance in three specific conditions, namely Wiedemann-Steiner (WDSTS), Kleefstra (KLEFS1) and Coffin-Siris syndrome (CSS1), with a particular focus on possible metabolic roots.Methods Eleven patients were enrolled, three with WDSTS, five with KLEFS1 and three with CSS1, referring to Fondazione IRCCS Ca’ Granda Ospedale Maggiore, Milan, Italy. We performed both a physical examination with detailed anthropometric measurements and an evaluation of the patients’ REE (rest energy expenditure) by indirect calorimetry, comparing the results with age- and sex-matched healthy controls.ResultsWe observed new clinical features and overlap between these conditions suggesting that different disturbances of epigenetic machinery genes can converge on a common effect, leading to overlapping clinical phenotypes. The REE was not distinguishable between the three conditions and healthy controls.ConclusionsEpigenetic machinery plays an essential role both in growth regulation and in neurodevelopment; we recommend evaluating skeletal [craniovertebral junction abnormalities (CVJ) polydactyly], otolaryngological [obstructive sleep apnea syndrome (OSAs), recurrent otitis media], dental [tooth agenesis, talon cusps], and central nervous system (CNS) [olfactory bulbs and cerebellum anomalies] features. These features could be included in monitoring guidelines. Further studies are needed to deepen the knowledge about energy metabolism.

Morphological comparison of astrocytes in the lamina cribrosa and glial lamina.

Although the mechanisms underlying glaucomatous neurodegeneration are not yet well understood, cellular and small animal models suggest that LC astrocytes undergo early morphologic and functional changes, indicating their role as early responders to glaucomatous stress. These models, however, lack the LC found in larger animals and humans, leaving the in situ morphology of LC astrocytes and their role in glaucoma initiation underexplored. In this work, we aimed to characterize the morphology of LC astrocytes in situ and determine differences and similarities with astrocytes in the mouse glial lamina (GL), the analogous structure in a prominent glaucoma model. Astrocytes in the LCs of twenty-two eyes from goats, sheep, and pigs were stochastically labeled via Multicolor DiOlistics and imaged in situ using confocal microscopy. 3D models of DiOlistically-labeled LC astrocytes and hGFAPpr-GFP mouse GL astrocytes were constructed to quantify morphological features related to astrocyte functions. LC and GL astrocyte cross-pore contacts, branching complexity, branch tortuosity, and cell and branch span were compared. LC astrocytes displayed distinct spatial relationships with collagen, greater branching complexity, and higher branch tortuosity compared to GL astrocytes. Despite substantial differences in their anatomical environments, LC and GL astrocytes had similar cell and branch spans. Astrocyte morphology in the LC was characterized through Multicolor DiOlistic labeling. LC and GL astrocytes have both distinct and shared morphological features. Further research is needed to understand the potentially unique roles of LC astrocytes in glaucoma initiation and progression.

Loan Defaults Prediction Based on Stacked Models Trained by Personalized Features

Training one machine learning model with features that all clients have will result in a waste of features, which is likely to adversely affect the model’s performance. To solve the problem, the study attempts a new method, which it to train an individual stacked model for each loan client based on personalised features. Data used contains information of about fifteen million loan applicants, their default status, and 468 features in all. 41 of the features that can be quantitively analysed are selected according to the feature importance output by a Random Forest model. Default prediction of every client is made by a stacked model trained with all selected features he/she has. The stacked model consists of two layers, in which a Light Gradient-Boosting Machine (LGBM) classifier is the base learner, and a Logistic Regression model is the meta learner. As the defaulters account for only 3.14%, which is significantly unbalanced, Area Under the Curve (AUC) and F1 scores are employed to evaluate the method, instead of accuracy. Test results show that models trained by personalised features perform better than the ones trained by shared features. Additionally, the stacked model outperforms individual Logistic Regression model, but performs nearly the same as individual LGBM Classifier. In detailed, the stacked models trained with personalised features result in AUC=0.772 and F1=0.188. Due to data unbalance, although the method’s F1 score is relatively low, it’s considered to be passable. In the future, stacked models combining different models will be attempted.

Multiscale Conditional Adversarial Networks based domain-adaptive method for rotating machinery fault diagnosis under variable working conditions

Deep learning has been increasingly used in health management and maintenance decision-making for rotating machinery. However, some challenges must be addressed to make this technology more effective. For example, the collected data is assumed to follow the same feature distribution, and sufficient labeled training data are available. Unfortunately, domain shifts occur inevitably in real-world scenarios due to different working conditions, and acquiring sufficient labeled samples is time-consuming and expensive in complex environments. This study proposes a novel domain adaptive framework called deep Multiscale Conditional Adversarial Networks (MCAN) for machinery fault diagnosis to address these shortcomings. The MCAN model comprises two key components. Constructed by a novel multiscale module with an attention mechanism, the first component is a shared feature generator that captures rich features at different internal perceptual scales, and the attention mechanism determines the weights assigned to each scale, enhancing the model’s dynamic adjustment and self-adaptation capabilities. The second component consists of two domain classifiers based on Bidirectional Long Short-Term Memory (BiLSTM) leveraging spatiotemporal features at various levels to achieve domain adaptation in the output space. The deep domain classifier also captures the cross-covariance dependencies between feature representations and classifier predictions, thereby improving the predictions’ discriminability. The proposed method has been evaluated using two publicly available fault diagnosis datasets and one condition monitoring experiment. The results of cross-domain transfer tasks demonstrated that the proposed method outperformed several state-of-the-art methods in terms of transferability and stability. This result is a significant step forward in deep learning for health management and maintenance decision-making for rotating machinery, and it has the potential to revolutionize its future application.

CRENet: Crowd region enhancement network for multi-person 3D pose estimation

Recovering multi-person 3D poses from a single image is a challenging problem due to inherent depth ambiguities, including root-relative depth and absolute root depth. Current bottom-up methods show promising potential to mitigate absolute root depth ambiguity through explicitly aggregating global contextual cues. However, these methods treat the entire image region equally during root depth regression, ignoring the negative impact of irrelevant regions. Moreover, they learn shared features for both depths, each of which focuses on different information. This sharing mechanism may result in negative transfer, thus diminishing root depth prediction accuracy. To address these challenges, we present a novel bottom-up method, Crowd Region Enhancement Network (CRENet), incorporating a Feature Decoupling Module (FDM) and a Global Attention Module (GAM). FDM explicitly learns the discriminative feature for each depth through adaptively recalibrating its channel-wise responses and fusing multi-level features, which makes the model focus on each depth prediction separately and thus avoids the adverse effect of negative transfer. GAM highlights crowd regions while suppressing irrelevant regions using the attention mechanism and further refines the attention regions based on the confidence measure about the attention, which is beneficial to learn depth-related cues from informative crowd regions and facilitate root depth estimation. Comprehensive experiments on benchmarks MuPoTS-3D and CMU Panoptic demonstrate that our method outperforms the state-of-the-art bottom-up methods in absolute 3D pose estimation and is applicable to in-the-wild images, which also indicates that learning depth-specific features and suppressing the noise signals can significantly benefit multi-person absolute 3D pose estimation.

Genomic insights of the emerging human pathogen Proteusappendicitidis sp. nov

ObjectivesProteus are known as opportunistic human pathogens that can cause a variety of infections. Proteus appendicitidis is a novel Proteus species associated with appendicitis, whereas their genomic characteristics and virulence potential remain understudied. This study aims to compare the genomic features of P. appendicitidis to that of the close Proteus species, and to assess its virulence-factor encoding capacity as an emerging pathogen. MethodsGenomes similar to that of P. appendicitidis HZ0627T were retrieved from the PATRIC-v3.6.10 web-server using the implanted Similar Genome Finder tool. Average nucleotide identity (ANI) between HZ0627T and the retrieved genomes was calculated using FastANI-v1.33. Core-genome sequences were extracted using Roary-v3.13.0, and core-genomic tree was constructed using FastTree-v2.1.11. Virulence-factor encoding capacity was predicted using PathoFact-v1.0. ResultsTwo previously unclassified Proteus sp. strains were reclassified as P. appendicitidis. Strains phylogenomically close to P. appendicitidis were clustered into five species, three of which were previously categorized under P. vulgaris biogroup 3. Remarkably, Proteus genomosp. 6 was identified as the closest species to P. appendicitidis, exhibiting ANI values ranging from 94.45 % to 94.95 % against HZ0627T. Genome annotation revealed shared genomic features and antimicrobial resistance (AMR) genes between P. appendicitidis and its phylogenetic neighbors. Additionally, P. appendicitidis is hypothesized to share infection mechanisms with Proteus genomosp. 6, as evidenced by the encoding of numerous virulence factors implicated in cell lysis and membrane pore-formation in the genome of both species. ConclusionsThis study provides genomic insights of P. appendicitidis sp. nov. and its taxonomic relatives, shedding light on their evolutionary relationships, pathogenic mechanisms, and AMR profiles. The findings are significant for the development of targeted therapeutic interventions against infections caused by this emerging pathogen.

Relation extraction for colorectal cancer via deep learning with entity-aware feature orthogonal decomposition

Relation extraction is significant for text structuring of colorectal cancer (CRC) pathological reports to facilitate doctors’ disease diagnoses. Although many relation extraction methods have been extensively studied for various natural language processing applications, they cannot be well transferred to be applied for CRC pathological reports since CRC pathological reports have some unique characteristics. To this end, a deep learning framework is designed in this paper to extract entity relations in CRC pathological reports, which is based on an encoder–decoder architecture with entity-aware feature orthogonal decomposition. Specifically, to effectively extract semantic features of long and short entities, a two-stream encoder is designed based on an edge-aware convolutional neural network and a dimension-aware dilated convolution residual network. To alleviate the influence of the blending of subject–object features, entity-aware feature orthogonal decomposition is designed to decompose the extracted semantic features into three types, i.e. subject features, object features and subject–object shared features. A stage-wise cross entropy loss is proposed to well train the network. Comparison experiments indicated that our designed network performs well on CRC pathological texts with the performance of 92.3% F1 score, 93.1% Precision, and 91.5% Recall, outperforming the existing relation extraction models.

Analyzing Wav2Vec 1.0 Embeddings for Cross-Database Parkinson's Disease Detection and Speech Features Extraction.

Advancements in deep learning speech representations have facilitated the effective use of extensive unlabeled speech datasets for Parkinson's disease (PD) modeling with minimal annotated data. This study employs the non-fine-tuned wav2vec 1.0 architecture to develop machine learning models for PD speech diagnosis tasks, such as cross-database classification and regression to predict demographic and articulation characteristics. The primary aim is to analyze overlapping components within the embeddings on both classification and regression tasks, investigating whether latent speech representations in PD are shared across models, particularly for related tasks. Firstly, evaluation using three multi-language PD datasets showed that wav2vec accurately detected PD based on speech, outperforming feature extraction using mel-frequency cepstral coefficients in the proposed cross-database classification scenarios. In cross-database scenarios using Italian and English-read texts, wav2vec demonstrated performance comparable to intra-dataset evaluations. We also compared our cross-database findings against those of other related studies. Secondly, wav2vec proved effective in regression, modeling various quantitative speech characteristics related to articulation and aging. Ultimately, subsequent analysis of important features examined the presence of significant overlaps between classification and regression models. The feature importance experiments discovered shared features across trained models, with increased sharing for related tasks, further suggesting that wav2vec contributes to improved generalizability. The study proposes wav2vec embeddings as a next promising step toward a speech-based universal model to assist in the evaluation of PD.

Catalytic Redundancies and Conformational Plasticity Drives Selectivity and Promiscuity in Quorum Quenching Lactonases.

Several enzymes from the metallo-β-lactamase-like family of lactonases (MLLs) degrade N-acyl L-homoserine lactones (AHLs). They play a role in a microbial communication system known as quorum sensing, which contributes to pathogenicity and biofilm formation. Designing quorum quenching (QQ) enzymes that can interfere with this communication allows them to be used in a range of industrial and biomedical applications. However, tailoring these enzymes for specific communication signals requires a thorough understanding of their mechanisms and the physicochemical properties that determine their substrate specificities. We present here a detailed biochemical, computational, and structural study of GcL, which is a highly proficient and thermostable MLL with broad substrate specificity. We show that GcL not only accepts a broad range of substrates but also hydrolyzes these substrates through at least two different mechanisms. Further, the preferred mechanism appears to depend on both the substrate structure and/or the nature of the residues lining the active site. We demonstrate that other lactonases, such as AiiA and AaL, show similar mechanistic promiscuity, suggesting that this is a shared feature among MLLs. Mechanistic promiscuity has been seen previously in the lactonase/paraoxonase PON1, as well as with protein tyrosine phosphatases that operate via a dual general acid mechanism. The apparent prevalence of this phenomenon is significant from both a biochemical and protein engineering perspective: in addition to optimizing for specific substrates, it may be possible to optimize for specific mechanisms, opening new doors not just for the design of novel quorum quenching enzymes but also of other mechanistically promiscuous enzymes.

Infections and their prognostic significance before diagnosis of chronic lymphocytic leukemia, non-Hodgkin lymphoma, or multiple myeloma

BackgroundImmunodeficiency is a shared feature of B cell malignancies. The risk of infections and their prognostic significance after diagnosis are well characterized, but, conversely, less is known about prediagnostic infections in these domains.MethodsIn matched case-control analyzes, using Danish nationwide registers, we assessed the rate of prediagnostic infections in chronic lymphocytic leukemia (CLL), diffuse large B cell lymphoma (DLBCL), multiple myeloma (MM), follicular lymphoma (FL), marginal zone lymphoma (MZL), and lymphoplasmacytic lymphoma (LPL). Survival analyzes of data from clinical registers were then used to determine the effect of infections in the year preceding diagnosis on overall survival. To yield results for as many patients as possible, antimicrobial prescriptions were used as surrogates for infections.ResultsThe nationwide and clinical registers comprised 30,389 patients, accumulating 213,649 antimicrobial prescriptions, and 18,560 patients accumulating 107,268 prescriptions, respectively. The relative risk of infections was increased up to 15 years prior to diagnosis of malignancy and markedly increased in the year just prior to diagnosis. More than two antimicrobials within one year prior to diagnosis were associated with significantly shorter overall survival, independently of known prognostic factors.ConclusionPatients with B cell-derived malignancies exhibit marked immunodeficiency several years prior to diagnosis such that different disease subtypes demonstrate both overlapping and distinct trends in infection risk preceding diagnosis. Moreover, multiple infections within the year preceding diagnosis are independently associated with shorter overall survival for all the examined malignancies.

Transformer- and joint learning-based dual-domain networks for undersampled MRI segmentation.

Recently, magnetic resonance imaging (MRI) has become a crucial medical imaging technology widely used in clinical practice. However, MRI faces challenges such as the lengthy acquisition time for k-space data and the need for time-consuming manual annotation by radiologists. Traditionally, these challenges have been addressed individually through undersampled MRI reconstruction and automatic segmentation algorithms. Whether undersampled MRI segmentation can be enhanced by treating undersampled MRI reconstruction and segmentation as an end-to-end task, trained simultaneously, rather than as serial tasks should beexplored. We introduce a novel Transformer- and Joint Learning-based Dual-domain Network (TJLD-Net) for undersampled MRIsegmentation. This method significantly enhances feature recognition in the segmentation process by fully utilizing the rich detail obtained during the image reconstruction phase. Consequently, the method can achieve precise and reliable image segmentation even with undersampled k-space data. Additionally, it incorporates an attention mechanism for feature enhancement, which improves the representation of shared features by learning the contextual information in MRimages. Simulation experiments demonstrate that the segmentation performance of TJLD-Net on three datasets is significantly higher than that of the joint model (RecSeg) and six baseline models (where reconstruction and segmentation are regarded as serial tasks). On the CHAOS dataset, the Dice scores of TJLD-Net are, on average, 9.87%, 2.17%, 1.90%, 1.80%, 9.60%, 0.80%, and 6.50% higher than those of the seven compared models. On the ATLAS challenge dataset, the average Dice scores of TJLD-Net improve by 4.23%, 5.63%, 2.30%, 1.53%, 3.57%, 0.93%, and 6.60%. Similarly, on the SKM-TEA dataset, the average Dice scores of TJLD-Net improve by 4.73%, 12.80%, 14.83%, 8.67%, 4.53%, 11.60%, and 12.10%. The novel TJLD-Net model provides a promising solution for undersampled MRI segmentation, overcoming the poor performance issues encountered by automated segmentation algorithms in low-quality acceleratedimaging.

Initial learning in the brain: From rules to action

Abstract We used fMRI to investigate the neural changes and representational dynamics associated with different learning modes during initial learning and subsequent implementation of previously acquired stimulus-response (S-R) associations. We compared instruction-based learning (INS) and trial-and-error learning (TE) via a third observation-based learning (OBS) condition. This was yoked to the TE condition and shared features with both, the INS and TE conditions. During learning, neural changes were observed in the Frontoparietal and Default Mode Networks across learning modes, consistent with a general decrease in cognitive control demand as learning progresses. INS and TE exhibited condition-specific signal changes, which we interpreted in the context of covert motor preparation during INS, and intentional action and increased cognitive control demand during early TE trials, respectively. Multivariate pattern analysis revealed individual rule information in bilateral prefrontal, premotor, and parietal cortices across learning modes. Most regions revealed consistent representations of individual S-R rules between the learning stage and subsequent implementation stage, regardless of the learning mode. This suggests that initially formed S-R rule representations guide task performance during S-R rule implementation, irrespective of how they are acquired. Finally, within the primary motor and sensory cortices, individual S-R rules were decodable during the learning stage not only when motor responses were overtly executed, as in TE, but also in the absence of overt motor execution, as in INS. This finding substantiates previous claims of covert motor preparatory mechanisms during INS.

Peripheral Blood Immune Cells from Individuals with Parkinson's Disease or Inflammatory Bowel Disease Share Deficits in Iron Storage and Transport that are Modulated by Non-Steroidal Anti-Inflammatory Drugs.

Parkinson's Disease (PD) is a multisystem disorder in which dysregulated neuroimmune crosstalk and inflammatory relay via the gut-blood-brain axis have been implicated in PD pathogenesis. Although alterations in circulating inflammatory cytokines and reactive oxygen species (ROS) have been associated with PD, no biomarkers have been identified that predict clinical progression or disease outcome. Gastrointestinal (GI) dysfunction, which involves perturbation of the underlying immune system, is an early and often-overlooked symptom that affects up to 80% of individuals living with PD. Interestingly, 50-70% of individuals with inflammatory bowel disease (IBD), a GI condition that has been epidemiologically linked to PD, display chronic illness-induced anemia - which drives toxic accumulation of iron in the gut. Ferroptotic (or iron loaded) cells have small and dysmorphic mitochondria-suggesting that mitochondrial dysfunction is a consequence of iron accumulation. In pro-inflammatory environments, iron accumulates in immune cells, suggesting a possible connection and/or synergy between iron dysregulation and immune cell dysfunction. Peripheral blood mononuclear cells (PBMCs) recapitulate certain PD-associated neuropathological and inflammatory signatures and can act as communicating messengers in the gut-brain axis. Additionally, this communication can be modulated by several environmental factors; specifically, our data further support existing literature demonstrating a role for non-steroidal anti-inflammatory drugs (NSAIDs) in modulating immune transcriptional states in inflamed individuals. A mechanism linking chronic gut inflammation to iron dysregulation and mitochondrial function within peripheral immune cells has yet to be identified in conferring risk for PD. To that end, we isolated PBMCs and simultaneously evaluated their directed transcriptome and bioenergetic status, to investigate if iron dysregulation and mitochondrial sensitization are linked in individuals living with PD or IBD because of chronic underlying remittent immune activation. We have identified shared features of peripheral inflammation and immunometabolism in individuals living with IBD or PD that may contribute to the epidemiological association reported between IBD and risk for PD.

Multi-modal data clustering using deep learning: A systematic review

Multi-modal clustering represents a formidable challenge in the domain of unsupervised learning. The objective of multi-modal clustering is to categorize data collected from diverse modalities, such as audio, visual, and textual sources, into distinct clusters. These clustering techniques operate by extracting shared features across modalities in an unsupervised manner, where the identified common features exhibit high correlations within real-world objects. Recognizing the importance of perceiving the correlated nature of these features is vital for enhancing clustering accuracy in multi-modal settings. This survey explores Deep Learning (DL) techniques applied to multi-modal clustering, encompassing methodologies such as Convolutional Neural Networks (CNN), Autoencoders (AE), Recurrent Neural Networks (RNN), and Graph Convolutional Networks (GCN). Notably, this survey represents the first attempt to investigate DL techniques specifically for multi-modal clustering. The survey presents a novel taxonomy for DL-based multi-modal clustering, conducts a comparative analysis of various multi-modal clustering approaches, and deliberates on the datasets employed in the evaluation process. Additionally, the survey identifies research gaps within the realm of multi-modal clustering, offering insights into potential future avenues for research.

Attack detection model for BCoT based on contrastive variational autoencoder and metric learning

With development of blockchain technology, clouding computing and Internet of Things (IoT), blockchain and cloud of things (BCoT) has become development tendency. But the security has become the most development hinder of BCoT. Attack detection model is a crucial part of attack revelation mechanism for BCoT. As a consequence, attack detection model has received more concerned. Due to the great diversity and variation of network attacks aiming to BCoT, tradition attack detection models are not suitable for BCoT. In this paper, we propose a novel attack detection model for BCoT, denoted as cVAE-DML. The novel model is based on contrastive variational autoencoder (cVAE) and deep metric learning (DML). By training the cVAE, the proposed model generates private features for attack traffic information as well as shared features between attack traffic information and normal traffic information. Based on those generated features, the proposed model can generate representative new samples to balance the training dataset. At last, the decoder of cVAE is connected to the deep metric learning network to detect attack aiming to BCoT. The efficiency of cVAE-DML is verified using the CIC-IDS 2017 dataset and CSE-CIC-IDS 2018 dataset. The results show that cVAE-DML can improve attack detection efficiency even under the condition of unbalanced samples.

Consistent and specific multi-view multi-label learning with correlation information

In multi-view multi-label (MVML) learning, each sample is represented by several heterogeneous distinct feature representations while associated with a set of class labels simultaneously. To achieve MVML learning, most of the existing methods contribute to the recovery of a consistent subspace, i.e., a shared feature representation, among multiple views. Nevertheless, each view has its inherent specific properties used in the discrimination process of labels. These methods lose sight in the specific information exploitation, and therefore are easily trapped in a sub-optimal result. In this study, we present an optimization framework CSVL to solve the learning problem. The main technical contribution in CSVL is a formulation for MVML learning while consistent subspace across views, specific subspace for each view, and the correlations among labels are taken into account. Specifically, consistent subspace is recovered by imposing a low-rank constraint among multiple views, and specific subspace of each view is extra generated with Frobenius norm. To further improve model generalization capability, we preserve both feature manifolds from multiple views and label correlations from multiple labels. Extensive experiments on 7 benchmark datasets show that our proposal CSVL has the advantages in MVML learning.

Structure-based pharmacophore modeling for precision inhibition of mutant ESR2 in breast cancer: A systematic computational approach.

Breast cancer, a leading cause of female mortality, is closely linked to mutations in estrogen receptor beta (ESR2), particularly in the ligand-binding domain, which contributed to altered signaling pathways and uncontrolled cell growth. This study investigates the molecular and structural aspects of ESR2 mutant proteins to identify shared pharmacophoric regions of ESR2 mutant proteins and potential therapeutic targets aligned within the pharmacophore model. This study was initiated by establishing a common pharmacophore model among three mutant ESR2 proteins (PDB ID: 2FSZ, 7XVZ, and 7XWR). The generated shared feature pharmacophore (SFP) includes four primary binding interactions: Hydrogen bond donors (HBD), hydrogen bond acceptors (HBA), hydrophobic interactions (HPho), and Aromatic interactions (Ar), along with halogen bond donors (XBD) and totalling 11 features (HBD: 2, HBA: 3, HPho: 3, Ar: 2, XBD: 1). By employing an in-house Python script, these 11 features distributed into 336 combinations, which were used as query to isolate a drug library of 41,248 compounds and subjected to virtual screening through the generated SFP. The virtual screening demonstrated 33 hits showing potential pharmacophoric fit scores and low RMSD value. The top four compounds: ZINC94272748, ZINC79046938, ZINC05925939, and ZINC59928516 showed a fit score of more than 86% and satisfied the Lipinski rule of five. These four compounds and a control underwent molecular (XP Glide mode) docking analysis against wild-type ESR2 protein (PDB ID: 1QKM), resulting in binding affinity of -8.26, -5.73, -10.80, and -8.42 kcal/mol, respectively, along with the control -7.2 kcal/mol. Furthermore, the stability of the selected candidates was determined through molecular dynamics (MD) simulations of 200 ns and MM-GBSA analysis. Based on MD simulations and MM-GBSA analysis, our study identified ZINC05925939 as a promising ESR2 inhibitor among the top four hits. However, it is essential to conduct further wet lab evaluation to assess its efficacy.