Traditional Chinese medicine(TCM)demonstrates distinctive advantages in disease prevention and treatment.However,analyzing its biological mechanisms through the modern medical research paradigm of“single drug,single ...Traditional Chinese medicine(TCM)demonstrates distinctive advantages in disease prevention and treatment.However,analyzing its biological mechanisms through the modern medical research paradigm of“single drug,single target”presents significant challenges due to its holistic approach.Network pharmacology and its core theory of network targets connect drugs and diseases from a holistic and systematic perspective based on biological networks,overcoming the limitations of reductionist research models and showing considerable value in TCM research.Recent integration of network target computational and experimental methods with artificial intelligence(AI)and multi-modal multi-omics technologies has substantially enhanced network pharmacology methodology.The advancement in computational and experimental techniques provides complementary support for network target theory in decoding TCM principles.This review,centered on network targets,examines the progress of network target methods combined with AI in predicting disease molecular mechanisms and drug-target relationships,alongside the application of multi-modal multi-omics technologies in analyzing TCM formulae,syndromes,and toxicity.Looking forward,network target theory is expected to incorporate emerging technologies while developing novel approaches aligned with its unique characteristics,potentially leading to significant breakthroughs in TCM research and advancing scientific understanding and innovation in TCM.展开更多
The rapid growth of biomedical data,particularly multi-omics data including genomes,transcriptomics,proteomics,metabolomics,and epigenomics,medical research and clinical decision-making confront both new opportunities...The rapid growth of biomedical data,particularly multi-omics data including genomes,transcriptomics,proteomics,metabolomics,and epigenomics,medical research and clinical decision-making confront both new opportunities and obstacles.The huge and diversified nature of these datasets cannot always be managed using traditional data analysis methods.As a consequence,deep learning has emerged as a strong tool for analysing numerous omics data due to its ability to handle complex and non-linear relationships.This paper explores the fundamental concepts of deep learning and how they are used in multi-omics medical data mining.We demonstrate how autoencoders,variational autoencoders,multimodal models,attention mechanisms,transformers,and graph neural networks enable pattern analysis and recognition across all omics data.Deep learning has been found to be effective in illness classification,biomarker identification,gene network learning,and therapeutic efficacy prediction.We also consider critical problems like as data quality,model explainability,whether findings can be repeated,and computational power requirements.We now consider future elements of combining omics with clinical and imaging data,explainable AI,federated learning,and real-time diagnostics.Overall,this study emphasises the need of collaborating across disciplines to advance deep learning-based multi-omics research for precision medicine and comprehending complicated disorders.展开更多
Objective To develop a non-invasive predictive model for coronary artery stenosis severity based on adaptive multi-modal integration of traditional Chinese and western medicine data.Methods Clinical indicators,echocar...Objective To develop a non-invasive predictive model for coronary artery stenosis severity based on adaptive multi-modal integration of traditional Chinese and western medicine data.Methods Clinical indicators,echocardiographic data,traditional Chinese medicine(TCM)tongue manifestations,and facial features were collected from patients who underwent coro-nary computed tomography angiography(CTA)in the Cardiac Care Unit(CCU)of Shanghai Tenth People's Hospital between May 1,2023 and May 1,2024.An adaptive weighted multi-modal data fusion(AWMDF)model based on deep learning was constructed to predict the severity of coronary artery stenosis.The model was evaluated using metrics including accura-cy,precision,recall,F1 score,and the area under the receiver operating characteristic(ROC)curve(AUC).Further performance assessment was conducted through comparisons with six ensemble machine learning methods,data ablation,model component ablation,and various decision-level fusion strategies.Results A total of 158 patients were included in the study.The AWMDF model achieved ex-cellent predictive performance(AUC=0.973,accuracy=0.937,precision=0.937,recall=0.929,and F1 score=0.933).Compared with model ablation,data ablation experiments,and various traditional machine learning models,the AWMDF model demonstrated superior per-formance.Moreover,the adaptive weighting strategy outperformed alternative approaches,including simple weighting,averaging,voting,and fixed-weight schemes.Conclusion The AWMDF model demonstrates potential clinical value in the non-invasive prediction of coronary artery disease and could serve as a tool for clinical decision support.展开更多
With the advent of the next-generation Air Traffic Control(ATC)system,there is growing interest in using Artificial Intelligence(AI)techniques to enhance Situation Awareness(SA)for ATC Controllers(ATCOs),i.e.,Intellig...With the advent of the next-generation Air Traffic Control(ATC)system,there is growing interest in using Artificial Intelligence(AI)techniques to enhance Situation Awareness(SA)for ATC Controllers(ATCOs),i.e.,Intelligent SA(ISA).However,the existing AI-based SA approaches often rely on unimodal data and lack a comprehensive description and benchmark of the ISA tasks utilizing multi-modal data for real-time ATC environments.To address this gap,by analyzing the situation awareness procedure of the ATCOs,the ISA task is refined to the processing of the two primary elements,i.e.,spoken instructions and flight trajectories.Subsequently,the ISA is further formulated into Controlling Intent Understanding(CIU)and Flight Trajectory Prediction(FTP)tasks.For the CIU task,an innovative automatic speech recognition and understanding framework is designed to extract the controlling intent from unstructured and continuous ATC communications.For the FTP task,the single-and multi-horizon FTP approaches are investigated to support the high-precision prediction of the situation evolution.A total of 32 unimodal/multi-modal advanced methods with extensive evaluation metrics are introduced to conduct the benchmarks on the real-world multi-modal ATC situation dataset.Experimental results demonstrate the effectiveness of AI-based techniques in enhancing ISA for the ATC environment.展开更多
Objective:Triple-negative breast cancer(TNBC)is a highly aggressive subtype that lacks targeted therapies,leading to a poorer prognosis.However,some patients achieve long-term recurrence-free survival(RFS),offering va...Objective:Triple-negative breast cancer(TNBC)is a highly aggressive subtype that lacks targeted therapies,leading to a poorer prognosis.However,some patients achieve long-term recurrence-free survival(RFS),offering valuable insights into tumor biology and potential treatment strategies.Methods:We conducted a comprehensive multi-omics analysis of 132 patients with American Joint Committee on Cancer(AJCC)stage III TNBC,comprising 36 long-term survivors(RFS≥8 years),62 moderate-term survivors(RFS:3-8 years),and 34 short-term survivors(RFS<3 years).Analyses investigated clinicopathological factors,whole-exome sequencing,germline mutations,copy number alterations(CNAs),RNA sequences,and metabolomic profiles.Results:Long-term survivors exhibited fewer metastatic regional lymph nodes,along with tumors showing reduced stromal fibrosis and lower Ki67 index.Molecularly,these tumors exhibited multiple alterations in genes related to homologous recombination repair,with higher frequencies of germline mutations and somatic CNAs.Additionally,tumors from long-term survivors demonstrated significant downregulation of the RTK-RAS signaling pathway.Metabolomic profiling revealed decreased levels of lipids and carbohydrate,particularly those involved in glycerophospholipid,fructose,and mannose metabolism,in long-term survival group.Multivariate Cox analysis identified fibrosis[hazard ratio(HR):12.70,95%confidence interval(95%CI):2.19-73.54,P=0.005]and RAC1copy number loss/deletion(HR:0.22,95%CI:0.06-0.83,P=0.026)as independent predictors of RFS.Higher fructose/mannose metabolism was associated with worse overall survival(HR:1.30,95%CI:1.01-1.68,P=0.045).Our findings emphasize the association between biological determinants and prolonged survival in patients with TNBC.Conclusions:Our study systematically identified the key molecular and metabolic features associated with prolonged survival in AJCC stage III TNBC,suggesting potential therapeutic targets to improve patient outcomes.展开更多
A personalized outfit recommendation has emerged as a hot research topic in the fashion domain.However,existing recommendations do not fully exploit user style preferences.Typically,users prefer particular styles such...A personalized outfit recommendation has emerged as a hot research topic in the fashion domain.However,existing recommendations do not fully exploit user style preferences.Typically,users prefer particular styles such as casual and athletic styles,and consider attributes like color and texture when selecting outfits.To achieve personalized outfit recommendations in line with user style preferences,this paper proposes a personal style guided outfit recommendation with multi-modal fashion compatibility modeling,termed as PSGNet.Firstly,a style classifier is designed to categorize fashion images of various clothing types and attributes into distinct style categories.Secondly,a personal style prediction module extracts user style preferences by analyzing historical data.Then,to address the limitations of single-modal representations and enhance fashion compatibility,both fashion images and text data are leveraged to extract multi-modal features.Finally,PSGNet integrates these components through Bayesian personalized ranking(BPR)to unify the personal style and fashion compatibility,where the former is used as personal style features and guides the output of the personalized outfit recommendation tailored to the target user.Extensive experiments on large-scale datasets demonstrate that the proposed model is efficient on the personalized outfit recommendation.展开更多
Lactic acid bacteria and the fermentation environment interact to form an intertwined system.Lactic acid bacteria are constantly evolving to adapt to different fermentation environments,causing changes in their physio...Lactic acid bacteria and the fermentation environment interact to form an intertwined system.Lactic acid bacteria are constantly evolving to adapt to different fermentation environments,causing changes in their physiological processes.To achieve a targeted improvement of their adaptability to various environments,a detail analysis of their evolutionary physiological processes is required.While several studies have been carried out in the past by using single-omics techniques to investigate their response to environmental stress,most researchers are now using a multi-omics approach to explore more detail in the biological regulatory networks and molecular mechanisms of lactic acid bacteria in response to environmental stress,thereby overcoming the limitations of single-omics analysis.In this review,we describe the various single-omics approaches that have been used to study environmental stress in lactic acid bacteria,present the advantages of various multi-omics combined analysis approaches,and discuss the potential and practicality of applying emerging single-cell transcriptomics and single-cell metabolomics techniques to the molecular mechanism study of microbes response to environmental stress.Multi-omics approaches enable the accurate identification of complex microbial physiological processes in different environments,allow people to comprehensively reveal the molecular mechanisms of microbes response to stress from different perspectives.Single-cell omics techniques,analyze the targeted regulation of microbial functions in a multi-dimensional space,provides a new perspective on understanding microbes responses environment stress.展开更多
Multi-modal Named Entity Recognition(MNER)aims to better identify meaningful textual entities by integrating information from images.Previous work has focused on extracting visual semantics at a fine-grained level,or ...Multi-modal Named Entity Recognition(MNER)aims to better identify meaningful textual entities by integrating information from images.Previous work has focused on extracting visual semantics at a fine-grained level,or obtaining entity related external knowledge from knowledge bases or Large Language Models(LLMs).However,these approaches ignore the poor semantic correlation between visual and textual modalities in MNER datasets and do not explore different multi-modal fusion approaches.In this paper,we present MMAVK,a multi-modal named entity recognition model with auxiliary visual knowledge and word-level fusion,which aims to leverage the Multi-modal Large Language Model(MLLM)as an implicit knowledge base.It also extracts vision-based auxiliary knowledge from the image formore accurate and effective recognition.Specifically,we propose vision-based auxiliary knowledge generation,which guides the MLLM to extract external knowledge exclusively derived from images to aid entity recognition by designing target-specific prompts,thus avoiding redundant recognition and cognitive confusion caused by the simultaneous processing of image-text pairs.Furthermore,we employ a word-level multi-modal fusion mechanism to fuse the extracted external knowledge with each word-embedding embedded from the transformerbased encoder.Extensive experimental results demonstrate that MMAVK outperforms or equals the state-of-the-art methods on the two classical MNER datasets,even when the largemodels employed have significantly fewer parameters than other baselines.展开更多
Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and ...Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and knowledge and the limitations of data sources,the visual knowledge within the knowledge graphs is generally of low quality,and some entities suffer from the issue of missing visual modality.Nevertheless,previous studies of MMKGC have primarily focused on how to facilitate modality interaction and fusion while neglecting the problems of low modality quality and modality missing.In this case,mainstream MMKGC models only use pre-trained visual encoders to extract features and transfer the semantic information to the joint embeddings through modal fusion,which inevitably suffers from problems such as error propagation and increased uncertainty.To address these problems,we propose a Multi-modal knowledge graph Completion model based on Super-resolution and Detailed Description Generation(MMCSD).Specifically,we leverage a pre-trained residual network to enhance the resolution and improve the quality of the visual modality.Moreover,we design multi-level visual semantic extraction and entity description generation,thereby further extracting entity semantics from structural triples and visual images.Meanwhile,we train a variational multi-modal auto-encoder and utilize a pre-trained multi-modal language model to complement the missing visual features.We conducted experiments on FB15K-237 and DB13K,and the results showed that MMCSD can effectively perform MMKGC and achieve state-of-the-art performance.展开更多
BACKGROUND Autoimmune liver diseases,including primary biliary cholangitis(PBC),autoi-mmune hepatitis(AIH),and their overlap syndrome(OS),involve immune-mediated liver injury,with OS occurring in 1.2%-25%of PBC patien...BACKGROUND Autoimmune liver diseases,including primary biliary cholangitis(PBC),autoi-mmune hepatitis(AIH),and their overlap syndrome(OS),involve immune-mediated liver injury,with OS occurring in 1.2%-25%of PBC patients.OS carries a higher risk of cirrhosis,hepatocellular carcinoma,and reduced survival.While its pathogenesis remains unclear,gut microbiota dysbiosis and serum metabolite alterations may play key roles.This study uses 16S rRNA sequencing and liquid chromatography-mass spec-trometry(LC-MS)metabolomics to compare gut microbiota and serum metabolites among PBC,AIH,and OS patients,and explores their associations with liver function.AIM To differentiate OS from PBC and AIH based on gut microbiota,serum metabolites,and liver function.METHODS Gut microbiota profiles were analyzed using 16S rRNA sequencing,while untargeted serum metabolomics was conducted via LC-MS.Comparative analyses were performed to identify differences in microbial composition and serum metabolite levels among PBC,AIH,and OS groups.Correlation analyses and network visualization tech-niques were applied to elucidate the interactions among liver function parameters,gut microbiota,and serum metabolites in OS patients.RESULTS Compared to patients with PBC or AIH,OS patients demonstrated significantly reduced microbial diversity and richness.Notable taxonomic shifts included decreased abundances of Firmicutes,Bacteroidetes,and Actinobacteria,alongside increased levels of Proteobacteria and Verrucomicrobia.Distinct serum metabolites,such as pentadecanoic acid and aminoimidazole carboxamide ribonucleotide,were identified in OS patients.Correlation analysis revealed that aspartate aminotransferase(AST)levels were negatively associated with the bacterial genus Fusicatenibacter and the metabolite L-Tyrosine.A microbial-metabolite network diagram further confirmed a strong association between Fusicatenibacter and L-Tyrosine in OS patients.CONCLUSION OS patients show decreased gut microbiota diversity and unique serum metabolites.Multi-omics linked AST,Fusicatenibacter,and L-Tyrosine,revealing OS mechanisms and diagnostic potential.展开更多
Objective To map the research hotspots,developmental trends,and existing challenges in the integration of artificial intelligence(AI)with multi-omics in traditional Chinese medicine(TCM)through comprehensive bibliomet...Objective To map the research hotspots,developmental trends,and existing challenges in the integration of artificial intelligence(AI)with multi-omics in traditional Chinese medicine(TCM)through comprehensive bibliometric analysis.Methods China National Knowledge Infrastructure(CNKI),Wanfang Data,China Science and Technology Journal Database(VIP),Chaoxing Journal Database,PubMed,and Web of Science were searched to collect literature on the theme of AI in TCM multi-omics research from the inception of each database to December 31,2024.Eligible records were required to simultaneously address AI,TCM,and multi-omics.Quantitative and visual analyses of publication growth,core authorship networks,institutional collaboration patterns,and keyword co-occurrence were performed using Microsoft Excel 2021,NoteExpress v4.0.0,and Cite-Space 6.3.R1.AI application modes in TCM multi-omics research were also categorized and summarized.Results A total of 1106 articles were enrolled(932 Chinese and 174 English).Publication output has increased continuously since 2010 and accelerated after 2016.Region-specific collaboration clusters were identified,dominated by Beijing University of Chinese Medicine,China Academy of Chinese Medical Sciences,Shanghai University of Traditional Chinese Medicine,and Nanjing University of Chinese Medicine.Keyword co-occurrence analysis revealed that current AI applications predominantly centered on metabolomics and algorithms such as cluster analysis and data mining.Research foci mainly ranked as follows:single herbs,herbal formulae,and disease-syndrome differentiation.Conclusion Machine learning methods are the predominant integrative modality of AI in the realm of TCM multi-omics research at present,utilized for processing omics data and uncovering latent patterns therein.The domain of TCM,in addition to investigating omics information procured through high-throughput technologies,also integrates data on traditional Chinese medicinal substances and clinical phenotypes,progressing towards joint analysis of multi-omics,high-dimensionality of data,and multi-modality of information.Deep learning approaches represent an emerging trend in the field.展开更多
Background Backfat thickness(BFT)is a vital economic trait in pigs,reflecting subcutaneous fat levels that affect meat quality and production efficiency.As a complex trait shaped by multiple genetic factors,BFT has be...Background Backfat thickness(BFT)is a vital economic trait in pigs,reflecting subcutaneous fat levels that affect meat quality and production efficiency.As a complex trait shaped by multiple genetic factors,BFT has been studied using genome-wide association studies(GWAS)and linkage analyses to locate fat-related quantitative trait loci(QTLs),but pinpointing causal variants and genes is hindered by linkage disequilibrium and limited regulatory data.This study aimed to dissect the QTLs affecting BFT on Sus scrofa chromosome 1(SSC1),elucidating regulatory variants,effector genes,and the cell types involved.Results Using whole-genome genotyping data from 3,578 pigs and phenotypic data for five BFT traits,we identified a 630.6 kb QTL on SSC1 significantly associated with these traits via GWAS and fine-mapping,pinpointing 34 candidate causal variants.Using deep convolutional neural networks to predict regulatory activity from sequence data integrated with detailed pig epigenetic profiles,we identified five SNPs potentially affecting enhancer activity in specific tissues.Notably,rs342950505(SSC1:161,123,588)influences weak enhancer activity across multiple tissues,including the brain.High-throughput chromosome conformation capture(Hi-C)analysis identified that rs342950505 interacts with eight genes.Chromatin state annotations confirmed enhancer activity at this QTL in the cerebellum.Leveraging these insights,single-cell ATAC-seq revealed a chromatin accessibility peak encompassing rs342950505 that regulates PMAIP1 expression in inhibitory neurons via enhancer-mediated mechanisms,with an adjacent peak modulating CCBE1 expression in neuroblasts and granule cells.Transcriptome-wide association studies(TWAS)confirmed PMAIP1's role in the hypothalamus,and Mendelian randomization(MR)validated PMAIP1 and CCBE1 as key brain expression quantitative trait locus(eQTL)effectors.We propose that the variant rs342950505,located within a regulatory peak,modulates PMAIP1 expression in inhibitory neurons,potentially influencing energy homeostasis via hypothalamic regulation.Similarly,CCBE1 may contribute to this process.Conclusions Our results,through systematic dissection of pleiotropic BFT-associated loci,provide a framework to elucidate regulatory mechanisms of complex traits,offering insights into polygenic control through lipid metabolism and neural signaling pathways.展开更多
Integrating multiple medical imaging techniques,including Magnetic Resonance Imaging(MRI),Computed Tomography,Positron Emission Tomography(PET),and ultrasound,provides a comprehensive view of the patient health status...Integrating multiple medical imaging techniques,including Magnetic Resonance Imaging(MRI),Computed Tomography,Positron Emission Tomography(PET),and ultrasound,provides a comprehensive view of the patient health status.Each of these methods contributes unique diagnostic insights,enhancing the overall assessment of patient condition.Nevertheless,the amalgamation of data from multiple modalities presents difficulties due to disparities in resolution,data collection methods,and noise levels.While traditional models like Convolutional Neural Networks(CNNs)excel in single-modality tasks,they struggle to handle multi-modal complexities,lacking the capacity to model global relationships.This research presents a novel approach for examining multi-modal medical imagery using a transformer-based system.The framework employs self-attention and cross-attention mechanisms to synchronize and integrate features across various modalities.Additionally,it shows resilience to variations in noise and image quality,making it adaptable for real-time clinical use.To address the computational hurdles linked to transformer models,particularly in real-time clinical applications in resource-constrained environments,several optimization techniques have been integrated to boost scalability and efficiency.Initially,a streamlined transformer architecture was adopted to minimize the computational load while maintaining model effectiveness.Methods such as model pruning,quantization,and knowledge distillation have been applied to reduce the parameter count and enhance the inference speed.Furthermore,efficient attention mechanisms such as linear or sparse attention were employed to alleviate the substantial memory and processing requirements of traditional self-attention operations.For further deployment optimization,researchers have implemented hardware-aware acceleration strategies,including the use of TensorRT and ONNX-based model compression,to ensure efficient execution on edge devices.These optimizations allow the approach to function effectively in real-time clinical settings,ensuring viability even in environments with limited resources.Future research directions include integrating non-imaging data to facilitate personalized treatment and enhancing computational efficiency for implementation in resource-limited environments.This study highlights the transformative potential of transformer models in multi-modal medical imaging,offering improvements in diagnostic accuracy and patient care outcomes.展开更多
Osmanthus fragrans Lour.is a well-known aromatic plant widely used as a food ingredient due to its unique floral fragrance and bioactive compounds.To fully utilize O.fragrans resources,we established an O.fragrans mul...Osmanthus fragrans Lour.is a well-known aromatic plant widely used as a food ingredient due to its unique floral fragrance and bioactive compounds.To fully utilize O.fragrans resources,we established an O.fragrans multi-omics database called the O.fragrans Information Resource(OfIR:http://yanglab.hzau.edu.cn/OfIR/home/).OfIR is a convenient and comprehensive multi-omics database that efficiently integrates phenotype and genetic variation from 127 O.fragrans cultivars,and provides many easy-to-use analysis tools,including primer design,sequence extraction,multi-sequence alignment,GO and KEGG enrichment analysis,variation annotation,and electronic PCR.Two case studies were used to demonstrate its power to mine candidate genetic variation sites or genes associated with specific traits or regulatory networks.In summary,the multi-omics database OfIR provides a convenient and user-friendly platform for researchers in mining functional genes and contributes to the genetic breeding of O.fragrans.展开更多
Background As an indigenous livestock species on the Tibetan Plateau,Tibetan sheep exhibit remarkable adaptability to low temperatures and nutrient-scarce environments.During the cold season,Tibetan sheep are typicall...Background As an indigenous livestock species on the Tibetan Plateau,Tibetan sheep exhibit remarkable adaptability to low temperatures and nutrient-scarce environments.During the cold season,Tibetan sheep are typically managed under two feeding regimes:barn feeding(BF)and traditional grazing(TG).However,the molecular mechanisms underlying their adaptation to these distinct management strategies remain unclear.This study aimed to investigate the adaptive strategies of rumen function in Tibetan sheep to cold-season feeding regimes by integrating analyses of rumen morphology,microbiome,metabolome,and transcriptome.Twelve healthy Tibetan sheep with similar body weights were assigned into two groups(BF vs.TG).At the end of the experiment,rumen tissues were subjected to histological observation.Multi-omics techniques were employed to evaluate the effects of cold-season feeding regimes on rumen function in Tibetan sheep.Results The ruminal papilla height,width,and muscular thickness were significantly higher in BF group.The relative abundances of Actinobacteria and Succiniclasticum were significantly elevated in the rumen of BF group,whereas Rikenellaceae,Gracilibacteria,and Lachnospiraceae showed higher abundances in the TG group.Metabolomic analysis identified 19 differential metabolites between the two groups,including upregulated compounds in BF group(fumaric acid,maltose,L-phenylalanine,and L-alanine)and TG group(e.g.,phenylacetic acid,salicyluric acid and ferulic acid).These metabolites were predominantly enriched in phenylalanine metabolism,alanine,aspartate and glutamate metabolism,and phenylalanine,tyrosine and tryptophan biosynthesis pathways.Additionally,210 differentially expressed genes(DEGs)were identified in rumen epithelium:100 upregulated DEGs in the BF group were enriched in nutrient metabolism-related pathways(e.g.,fatty acid degradation and PPAR signaling pathway),while 110 upregulated DEGs in the TG group were associated with immune-related pathways(e.g.,p53 signaling pathway and glutathione metabolism).Conclusions Among these,we observed distinct rumen functional responses to different cold-season feeding regimes in Tibetan sheep and revealed energy allocation strategies mediated by host-microbe interactions.In the BF group,Tibetan sheep adopted a"metabolic efficiency-priority"strategy,driving rumen microbiota to maximize energy capture from high-nutrient diets to support host growth.In contrast,the TG group exhibited an"environmental adaptation-priority"strategy,where rumen microbiota prioritized cellulose degradation and anti-inflammatory functions,reallocating energy toward homeostasis maintenance at the expense of rumen development and growth performance.展开更多
BACKGROUND Gastrointestinal(GI)malignancies,including gastric and colorectal cancers,remain one of the primary contributors to cancer-related illness and death globally.Despite the availability of conventional diagnos...BACKGROUND Gastrointestinal(GI)malignancies,including gastric and colorectal cancers,remain one of the primary contributors to cancer-related illness and death globally.Despite the availability of conventional diagnostic tools,early detection and personalized treatment remain significant clinical challenges.Integrated multi-omics methods encompassing genomic,transcriptomic,proteomic,metabolomic,and microbiome profiles have emerged as powerful tools for advancing precision oncology,improving diagnostic accuracy,and informing therapeutic strategies.AIM To investigate the application of multi-omics approaches in the early detection,risk stratification,treatment optimization,and biomarker discovery of GI malignancies.METHODS The systematic review process was conducted in accordance with the PRISMA 2020 guidelines.Five databases,PubMed,ScienceDirect,Scopus,ProQuest,and Web of Science,were searched for studies published in English from 2015 onwards.Eligible studies involved human subjects and focused on multi-omics integration in GI cancers,including biomarker identification,tumor microenvironment analysis,tumor heterogeneity,organoid modeling,and artificial intelligence(AI)-driven analytics.Data extraction included study characteristics,omics modalities,clinical applications,and evaluation of study quality conducted with the Cochrane risk of bias 2.0 instrument.RESULTS A total of 17196 initially identified articles,20 met the inclusion criteria.The findings highlight the superiority of multi-omics platforms over traditional biomarkers(e.g.,carcinoembryonic antigen and carbohydrate antigen 19-9 in detecting early stage GI cancers.Key applications include the identification of circulating tumor DNA,extracellular vesicles,lipidomic and proteomic signatures,and the adoption of AI algorithms to enhance diagnostic precision.Multi-omics analysis has also revealed the mechanisms of immune modulation,tumor microenvironment regulation,metastatic behavior,and drug resistance.Organoid models and microbiota profiling have contributed to personalized therapeutic strategies and immunotherapy optimization.CONCLUSION Multi-omics approaches offer significant advancements in the early diagnosis,prognostic evaluation,and personalized treatment of GI malignancies.Their integration with AI analytics,organoid biobanking,and microbiota modulation provides a pathway for precision oncology research.展开更多
In this editorial,we discuss the findings reported by Wang et al in the latest issue of the World Journal of Gastrointestinal Oncology.Various research methodologies,including microbiome analysis,assert that the Tzu-C...In this editorial,we discuss the findings reported by Wang et al in the latest issue of the World Journal of Gastrointestinal Oncology.Various research methodologies,including microbiome analysis,assert that the Tzu-Chi Cancer-Antagonizing and Life-Protecting II Decoction of Chinese herbal compounds mitigates inflammatory responses by inhibiting the NF-κB signaling pathway.This action helps maintain the dynamic equilibrium of the intestinal microecology and lessens chemotherapy-induced gastrointestinal damage.The efficacy of these compounds is intimately linked to the composition of intestinal microbes.These compounds regulate intestinal microecology by virtue of their specific compatibility and effectiveness,thereby enhancing the overall therapeutic outcomes of cancer chemotherapy.Nonetheless,the exact mechanisms underlying these effects warrant further investigation.Multi-omics technologies offer a systematic approach to elucidate the mechanisms and effectiveness of Chinese herbal compounds in vivo.This manuscript reviews the application of multi-omics technologies to Chinese herbal compounds and explores their potential role in modulating the gastrointestinal microenvironment following cancer chemotherapy,thus providing a theoretical foundation for their continued use in adjunct cancer treatment.展开更多
Objective Pneumoconiosis,a lung disease caused by irreversible fibrosis,represents a significant public health burden.This study investigates the causal relationships between gut microbiota,gene methylation,gene expre...Objective Pneumoconiosis,a lung disease caused by irreversible fibrosis,represents a significant public health burden.This study investigates the causal relationships between gut microbiota,gene methylation,gene expression,protein levels,and pneumoconiosis using a multi-omics approach and Mendelian randomization(MR).Methods We analyzed gut microbiota data from MiBioGen and Esteban et al.to assess their potential causal effects on pneumoconiosis subtypes(asbestosis,silicosis,and inorganic pneumoconiosis)using conventional and summary-data-based MR(SMR).Gene methylation and expression data from Genotype-Tissue Expression and eQTLGen,along with protein level data from deCODE and UK Biobank Pharma Proteomics Project,were examined in relation to pneumoconiosis data from FinnGen.To validate our findings,we assessed self-measured gut flora from a pneumoconiosis cohort and performed fine mapping,drug prediction,molecular docking,and Phenome-Wide Association Studies to explore relevant phenotypes of key genes.Results Three core gut microorganisms were identified:Romboutsia(OR=0.249)as a protective factor against silicosis,Pasteurellaceae(OR=3.207)and Haemophilus parainfluenzae(OR=2.343)as risk factors for inorganic pneumoconiosis.Additionally,mapping and quantitative trait loci analyses revealed that the genes VIM,STX8,and MIF were significantly associated with pneumoconiosis risk.Conclusions This multi-omics study highlights the associations between gut microbiota and key genes(VIM,STX8,MIF)with pneumoconiosis,offering insights into potential therapeutic targets and personalized treatment strategies.展开更多
To address the challenge of missing modal information in entity alignment and to mitigate information loss or bias arising frommodal heterogeneity during fusion,while also capturing shared information acrossmodalities...To address the challenge of missing modal information in entity alignment and to mitigate information loss or bias arising frommodal heterogeneity during fusion,while also capturing shared information acrossmodalities,this paper proposes a Multi-modal Pre-synergistic Entity Alignmentmodel based on Cross-modalMutual Information Strategy Optimization(MPSEA).The model first employs independent encoders to process multi-modal features,including text,images,and numerical values.Next,a multi-modal pre-synergistic fusion mechanism integrates graph structural and visual modal features into the textual modality as preparatory information.This pre-fusion strategy enables unified perception of heterogeneous modalities at the model’s initial stage,reducing discrepancies during the fusion process.Finally,using cross-modal deep perception reinforcement learning,the model achieves adaptive multilevel feature fusion between modalities,supporting learningmore effective alignment strategies.Extensive experiments on multiple public datasets show that the MPSEA method achieves gains of up to 7% in Hits@1 and 8.2% in MRR on the FBDB15K dataset,and up to 9.1% in Hits@1 and 7.7% in MRR on the FBYG15K dataset,compared to existing state-of-the-art methods.These results confirm the effectiveness of the proposed model.展开更多
Cancer rates are increasing globally,making it more urgent than ever to enhance research and treatment strategies.This study aims to investigate how innovative technology and integrated multi-omics techniques could he...Cancer rates are increasing globally,making it more urgent than ever to enhance research and treatment strategies.This study aims to investigate how innovative technology and integrated multi-omics techniques could help improve cancer diagnosis,knowledge,and therapy.A complete literature search was undertaken using PubMed,Elsevier,Google Scholar,ScienceDirect,Embase,and NCBI.This review examined the articles published from 2010 to 2025.Relevant articles were found using keywords and selected using inclusion criteria New sequencing methods,like next-generation sequencing and single-cell analysis,have transformed our ability to study tumor complexity and genetic mutations,paving the way for more precise,personalized treatments.At the same time,imaging technologies such as Positron Emission Tomography(PET)and Magnetic Resonance Imaging(MRI)have made detecting tumors early and tracking treatment progress easier,all while improving patient comfort.Artificial intelligence(AI)and machine learning(ML)are having a significant impact by helping to analyze large volumes of data more efficiently and enhancing diagnostic accuracy.Meanwhile,Clustered Regulatory Interspaced Short Palindromic Repeats(CRISPR/Cas9)gene editing is emerging as a promising tool for directly targeting genes related to cancer,providing new possibilities for treatment.By integrating genomic,transcriptomic,proteomic,and metabolomic data,multi-omics approaches provide researchers with a more comprehensive understanding of the molecular mechanisms driving cancer,thereby facilitating the discovery of novel biomarkers and therapeutic targets.Despite these advancements,additional challenges persist,such as data integration,elevated costs,standardisation concerns,and the intricacies of translating findings into clinical practice,which might prevent wider implementation.Research needs to concentrate on improving these developments and encouraging multidisciplinary cooperation going forward to maximize their possibilities.Personalized cancer therapies will become more successful with ongoing developments,therefore enhancing patient outcomes and quality of life.展开更多
文摘Traditional Chinese medicine(TCM)demonstrates distinctive advantages in disease prevention and treatment.However,analyzing its biological mechanisms through the modern medical research paradigm of“single drug,single target”presents significant challenges due to its holistic approach.Network pharmacology and its core theory of network targets connect drugs and diseases from a holistic and systematic perspective based on biological networks,overcoming the limitations of reductionist research models and showing considerable value in TCM research.Recent integration of network target computational and experimental methods with artificial intelligence(AI)and multi-modal multi-omics technologies has substantially enhanced network pharmacology methodology.The advancement in computational and experimental techniques provides complementary support for network target theory in decoding TCM principles.This review,centered on network targets,examines the progress of network target methods combined with AI in predicting disease molecular mechanisms and drug-target relationships,alongside the application of multi-modal multi-omics technologies in analyzing TCM formulae,syndromes,and toxicity.Looking forward,network target theory is expected to incorporate emerging technologies while developing novel approaches aligned with its unique characteristics,potentially leading to significant breakthroughs in TCM research and advancing scientific understanding and innovation in TCM.
文摘The rapid growth of biomedical data,particularly multi-omics data including genomes,transcriptomics,proteomics,metabolomics,and epigenomics,medical research and clinical decision-making confront both new opportunities and obstacles.The huge and diversified nature of these datasets cannot always be managed using traditional data analysis methods.As a consequence,deep learning has emerged as a strong tool for analysing numerous omics data due to its ability to handle complex and non-linear relationships.This paper explores the fundamental concepts of deep learning and how they are used in multi-omics medical data mining.We demonstrate how autoencoders,variational autoencoders,multimodal models,attention mechanisms,transformers,and graph neural networks enable pattern analysis and recognition across all omics data.Deep learning has been found to be effective in illness classification,biomarker identification,gene network learning,and therapeutic efficacy prediction.We also consider critical problems like as data quality,model explainability,whether findings can be repeated,and computational power requirements.We now consider future elements of combining omics with clinical and imaging data,explainable AI,federated learning,and real-time diagnostics.Overall,this study emphasises the need of collaborating across disciplines to advance deep learning-based multi-omics research for precision medicine and comprehending complicated disorders.
基金Construction Program of the Key Discipline of State Administration of Traditional Chinese Medicine of China(ZYYZDXK-2023069)Research Project of Shanghai Municipal Health Commission (2024QN018)Shanghai University of Traditional Chinese Medicine Science and Technology Development Program (23KFL005)。
文摘Objective To develop a non-invasive predictive model for coronary artery stenosis severity based on adaptive multi-modal integration of traditional Chinese and western medicine data.Methods Clinical indicators,echocardiographic data,traditional Chinese medicine(TCM)tongue manifestations,and facial features were collected from patients who underwent coro-nary computed tomography angiography(CTA)in the Cardiac Care Unit(CCU)of Shanghai Tenth People's Hospital between May 1,2023 and May 1,2024.An adaptive weighted multi-modal data fusion(AWMDF)model based on deep learning was constructed to predict the severity of coronary artery stenosis.The model was evaluated using metrics including accura-cy,precision,recall,F1 score,and the area under the receiver operating characteristic(ROC)curve(AUC).Further performance assessment was conducted through comparisons with six ensemble machine learning methods,data ablation,model component ablation,and various decision-level fusion strategies.Results A total of 158 patients were included in the study.The AWMDF model achieved ex-cellent predictive performance(AUC=0.973,accuracy=0.937,precision=0.937,recall=0.929,and F1 score=0.933).Compared with model ablation,data ablation experiments,and various traditional machine learning models,the AWMDF model demonstrated superior per-formance.Moreover,the adaptive weighting strategy outperformed alternative approaches,including simple weighting,averaging,voting,and fixed-weight schemes.Conclusion The AWMDF model demonstrates potential clinical value in the non-invasive prediction of coronary artery disease and could serve as a tool for clinical decision support.
基金supported by the National Natural Science Foundation of China(Nos.62371323,62401380,U2433217,U2333209,and U20A20161)Natural Science Foundation of Sichuan Province,China(Nos.2025ZNSFSC1476)+2 种基金Sichuan Science and Technology Program,China(Nos.2024YFG0010 and 2024ZDZX0046)the Institutional Research Fund from Sichuan University(Nos.2024SCUQJTX030)the Open Fund of Key Laboratory of Flight Techniques and Flight Safety,CAAC(Nos.GY2024-01A).
文摘With the advent of the next-generation Air Traffic Control(ATC)system,there is growing interest in using Artificial Intelligence(AI)techniques to enhance Situation Awareness(SA)for ATC Controllers(ATCOs),i.e.,Intelligent SA(ISA).However,the existing AI-based SA approaches often rely on unimodal data and lack a comprehensive description and benchmark of the ISA tasks utilizing multi-modal data for real-time ATC environments.To address this gap,by analyzing the situation awareness procedure of the ATCOs,the ISA task is refined to the processing of the two primary elements,i.e.,spoken instructions and flight trajectories.Subsequently,the ISA is further formulated into Controlling Intent Understanding(CIU)and Flight Trajectory Prediction(FTP)tasks.For the CIU task,an innovative automatic speech recognition and understanding framework is designed to extract the controlling intent from unstructured and continuous ATC communications.For the FTP task,the single-and multi-horizon FTP approaches are investigated to support the high-precision prediction of the situation evolution.A total of 32 unimodal/multi-modal advanced methods with extensive evaluation metrics are introduced to conduct the benchmarks on the real-world multi-modal ATC situation dataset.Experimental results demonstrate the effectiveness of AI-based techniques in enhancing ISA for the ATC environment.
基金supported by grants from the Medical Engineering Jiont Fund of the Fudan University(No.IDH2310117)。
文摘Objective:Triple-negative breast cancer(TNBC)is a highly aggressive subtype that lacks targeted therapies,leading to a poorer prognosis.However,some patients achieve long-term recurrence-free survival(RFS),offering valuable insights into tumor biology and potential treatment strategies.Methods:We conducted a comprehensive multi-omics analysis of 132 patients with American Joint Committee on Cancer(AJCC)stage III TNBC,comprising 36 long-term survivors(RFS≥8 years),62 moderate-term survivors(RFS:3-8 years),and 34 short-term survivors(RFS<3 years).Analyses investigated clinicopathological factors,whole-exome sequencing,germline mutations,copy number alterations(CNAs),RNA sequences,and metabolomic profiles.Results:Long-term survivors exhibited fewer metastatic regional lymph nodes,along with tumors showing reduced stromal fibrosis and lower Ki67 index.Molecularly,these tumors exhibited multiple alterations in genes related to homologous recombination repair,with higher frequencies of germline mutations and somatic CNAs.Additionally,tumors from long-term survivors demonstrated significant downregulation of the RTK-RAS signaling pathway.Metabolomic profiling revealed decreased levels of lipids and carbohydrate,particularly those involved in glycerophospholipid,fructose,and mannose metabolism,in long-term survival group.Multivariate Cox analysis identified fibrosis[hazard ratio(HR):12.70,95%confidence interval(95%CI):2.19-73.54,P=0.005]and RAC1copy number loss/deletion(HR:0.22,95%CI:0.06-0.83,P=0.026)as independent predictors of RFS.Higher fructose/mannose metabolism was associated with worse overall survival(HR:1.30,95%CI:1.01-1.68,P=0.045).Our findings emphasize the association between biological determinants and prolonged survival in patients with TNBC.Conclusions:Our study systematically identified the key molecular and metabolic features associated with prolonged survival in AJCC stage III TNBC,suggesting potential therapeutic targets to improve patient outcomes.
基金Shanghai Frontier Science Research Center for Modern Textiles,Donghua University,ChinaOpen Project of Henan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment,Zhengzhou University of Light Industry,China(No.IM202303)National Key Research and Development Program of China(No.2019YFB1706300)。
文摘A personalized outfit recommendation has emerged as a hot research topic in the fashion domain.However,existing recommendations do not fully exploit user style preferences.Typically,users prefer particular styles such as casual and athletic styles,and consider attributes like color and texture when selecting outfits.To achieve personalized outfit recommendations in line with user style preferences,this paper proposes a personal style guided outfit recommendation with multi-modal fashion compatibility modeling,termed as PSGNet.Firstly,a style classifier is designed to categorize fashion images of various clothing types and attributes into distinct style categories.Secondly,a personal style prediction module extracts user style preferences by analyzing historical data.Then,to address the limitations of single-modal representations and enhance fashion compatibility,both fashion images and text data are leveraged to extract multi-modal features.Finally,PSGNet integrates these components through Bayesian personalized ranking(BPR)to unify the personal style and fashion compatibility,where the former is used as personal style features and guides the output of the personalized outfit recommendation tailored to the target user.Extensive experiments on large-scale datasets demonstrate that the proposed model is efficient on the personalized outfit recommendation.
基金supported by the National Natural Science Foundation of China(32160578)the Ningxia Hui Autonomous Region Key Research and Develoment Program(2023BCF01027).
文摘Lactic acid bacteria and the fermentation environment interact to form an intertwined system.Lactic acid bacteria are constantly evolving to adapt to different fermentation environments,causing changes in their physiological processes.To achieve a targeted improvement of their adaptability to various environments,a detail analysis of their evolutionary physiological processes is required.While several studies have been carried out in the past by using single-omics techniques to investigate their response to environmental stress,most researchers are now using a multi-omics approach to explore more detail in the biological regulatory networks and molecular mechanisms of lactic acid bacteria in response to environmental stress,thereby overcoming the limitations of single-omics analysis.In this review,we describe the various single-omics approaches that have been used to study environmental stress in lactic acid bacteria,present the advantages of various multi-omics combined analysis approaches,and discuss the potential and practicality of applying emerging single-cell transcriptomics and single-cell metabolomics techniques to the molecular mechanism study of microbes response to environmental stress.Multi-omics approaches enable the accurate identification of complex microbial physiological processes in different environments,allow people to comprehensively reveal the molecular mechanisms of microbes response to stress from different perspectives.Single-cell omics techniques,analyze the targeted regulation of microbial functions in a multi-dimensional space,provides a new perspective on understanding microbes responses environment stress.
基金funded by Research Project,grant number BHQ090003000X03.
文摘Multi-modal Named Entity Recognition(MNER)aims to better identify meaningful textual entities by integrating information from images.Previous work has focused on extracting visual semantics at a fine-grained level,or obtaining entity related external knowledge from knowledge bases or Large Language Models(LLMs).However,these approaches ignore the poor semantic correlation between visual and textual modalities in MNER datasets and do not explore different multi-modal fusion approaches.In this paper,we present MMAVK,a multi-modal named entity recognition model with auxiliary visual knowledge and word-level fusion,which aims to leverage the Multi-modal Large Language Model(MLLM)as an implicit knowledge base.It also extracts vision-based auxiliary knowledge from the image formore accurate and effective recognition.Specifically,we propose vision-based auxiliary knowledge generation,which guides the MLLM to extract external knowledge exclusively derived from images to aid entity recognition by designing target-specific prompts,thus avoiding redundant recognition and cognitive confusion caused by the simultaneous processing of image-text pairs.Furthermore,we employ a word-level multi-modal fusion mechanism to fuse the extracted external knowledge with each word-embedding embedded from the transformerbased encoder.Extensive experimental results demonstrate that MMAVK outperforms or equals the state-of-the-art methods on the two classical MNER datasets,even when the largemodels employed have significantly fewer parameters than other baselines.
基金funded by Research Project,grant number BHQ090003000X03。
文摘Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and knowledge and the limitations of data sources,the visual knowledge within the knowledge graphs is generally of low quality,and some entities suffer from the issue of missing visual modality.Nevertheless,previous studies of MMKGC have primarily focused on how to facilitate modality interaction and fusion while neglecting the problems of low modality quality and modality missing.In this case,mainstream MMKGC models only use pre-trained visual encoders to extract features and transfer the semantic information to the joint embeddings through modal fusion,which inevitably suffers from problems such as error propagation and increased uncertainty.To address these problems,we propose a Multi-modal knowledge graph Completion model based on Super-resolution and Detailed Description Generation(MMCSD).Specifically,we leverage a pre-trained residual network to enhance the resolution and improve the quality of the visual modality.Moreover,we design multi-level visual semantic extraction and entity description generation,thereby further extracting entity semantics from structural triples and visual images.Meanwhile,we train a variational multi-modal auto-encoder and utilize a pre-trained multi-modal language model to complement the missing visual features.We conducted experiments on FB15K-237 and DB13K,and the results showed that MMCSD can effectively perform MMKGC and achieve state-of-the-art performance.
基金Supported by WBE Liver Foundation,No.WBE20220182022 Young and Middle-aged Talents Incubation Project(Youth Innovation)of Beijing Youan Hospital,Capital Medical University,No.BJYAYY-YN-2022-092023 Young and Middle-aged Talents Incubation Project(Youth Innovation)of Beijing Youan Hospital,Capital Medical University,No.BJYAYYYN2023-14.
文摘BACKGROUND Autoimmune liver diseases,including primary biliary cholangitis(PBC),autoi-mmune hepatitis(AIH),and their overlap syndrome(OS),involve immune-mediated liver injury,with OS occurring in 1.2%-25%of PBC patients.OS carries a higher risk of cirrhosis,hepatocellular carcinoma,and reduced survival.While its pathogenesis remains unclear,gut microbiota dysbiosis and serum metabolite alterations may play key roles.This study uses 16S rRNA sequencing and liquid chromatography-mass spec-trometry(LC-MS)metabolomics to compare gut microbiota and serum metabolites among PBC,AIH,and OS patients,and explores their associations with liver function.AIM To differentiate OS from PBC and AIH based on gut microbiota,serum metabolites,and liver function.METHODS Gut microbiota profiles were analyzed using 16S rRNA sequencing,while untargeted serum metabolomics was conducted via LC-MS.Comparative analyses were performed to identify differences in microbial composition and serum metabolite levels among PBC,AIH,and OS groups.Correlation analyses and network visualization tech-niques were applied to elucidate the interactions among liver function parameters,gut microbiota,and serum metabolites in OS patients.RESULTS Compared to patients with PBC or AIH,OS patients demonstrated significantly reduced microbial diversity and richness.Notable taxonomic shifts included decreased abundances of Firmicutes,Bacteroidetes,and Actinobacteria,alongside increased levels of Proteobacteria and Verrucomicrobia.Distinct serum metabolites,such as pentadecanoic acid and aminoimidazole carboxamide ribonucleotide,were identified in OS patients.Correlation analysis revealed that aspartate aminotransferase(AST)levels were negatively associated with the bacterial genus Fusicatenibacter and the metabolite L-Tyrosine.A microbial-metabolite network diagram further confirmed a strong association between Fusicatenibacter and L-Tyrosine in OS patients.CONCLUSION OS patients show decreased gut microbiota diversity and unique serum metabolites.Multi-omics linked AST,Fusicatenibacter,and L-Tyrosine,revealing OS mechanisms and diagnostic potential.
基金General Project of Scientific Research of Hunan Provincial Education Department (22C0191)General Project of University-level Scientific Research of Hunan University of Chinese Medicine (Z2023XJYB21)Hunan Provincial Degree and Graduate Education Reform Research Project(2024JGYB157)。
文摘Objective To map the research hotspots,developmental trends,and existing challenges in the integration of artificial intelligence(AI)with multi-omics in traditional Chinese medicine(TCM)through comprehensive bibliometric analysis.Methods China National Knowledge Infrastructure(CNKI),Wanfang Data,China Science and Technology Journal Database(VIP),Chaoxing Journal Database,PubMed,and Web of Science were searched to collect literature on the theme of AI in TCM multi-omics research from the inception of each database to December 31,2024.Eligible records were required to simultaneously address AI,TCM,and multi-omics.Quantitative and visual analyses of publication growth,core authorship networks,institutional collaboration patterns,and keyword co-occurrence were performed using Microsoft Excel 2021,NoteExpress v4.0.0,and Cite-Space 6.3.R1.AI application modes in TCM multi-omics research were also categorized and summarized.Results A total of 1106 articles were enrolled(932 Chinese and 174 English).Publication output has increased continuously since 2010 and accelerated after 2016.Region-specific collaboration clusters were identified,dominated by Beijing University of Chinese Medicine,China Academy of Chinese Medical Sciences,Shanghai University of Traditional Chinese Medicine,and Nanjing University of Chinese Medicine.Keyword co-occurrence analysis revealed that current AI applications predominantly centered on metabolomics and algorithms such as cluster analysis and data mining.Research foci mainly ranked as follows:single herbs,herbal formulae,and disease-syndrome differentiation.Conclusion Machine learning methods are the predominant integrative modality of AI in the realm of TCM multi-omics research at present,utilized for processing omics data and uncovering latent patterns therein.The domain of TCM,in addition to investigating omics information procured through high-throughput technologies,also integrates data on traditional Chinese medicinal substances and clinical phenotypes,progressing towards joint analysis of multi-omics,high-dimensionality of data,and multi-modality of information.Deep learning approaches represent an emerging trend in the field.
基金supported by the China Postdoctoral Science Foundation[Grant Number BX20240146 and 2024M761230]Key Project of Research and Development Plan in Jiangxi Province[Grant Number 20243BCC31001].
文摘Background Backfat thickness(BFT)is a vital economic trait in pigs,reflecting subcutaneous fat levels that affect meat quality and production efficiency.As a complex trait shaped by multiple genetic factors,BFT has been studied using genome-wide association studies(GWAS)and linkage analyses to locate fat-related quantitative trait loci(QTLs),but pinpointing causal variants and genes is hindered by linkage disequilibrium and limited regulatory data.This study aimed to dissect the QTLs affecting BFT on Sus scrofa chromosome 1(SSC1),elucidating regulatory variants,effector genes,and the cell types involved.Results Using whole-genome genotyping data from 3,578 pigs and phenotypic data for five BFT traits,we identified a 630.6 kb QTL on SSC1 significantly associated with these traits via GWAS and fine-mapping,pinpointing 34 candidate causal variants.Using deep convolutional neural networks to predict regulatory activity from sequence data integrated with detailed pig epigenetic profiles,we identified five SNPs potentially affecting enhancer activity in specific tissues.Notably,rs342950505(SSC1:161,123,588)influences weak enhancer activity across multiple tissues,including the brain.High-throughput chromosome conformation capture(Hi-C)analysis identified that rs342950505 interacts with eight genes.Chromatin state annotations confirmed enhancer activity at this QTL in the cerebellum.Leveraging these insights,single-cell ATAC-seq revealed a chromatin accessibility peak encompassing rs342950505 that regulates PMAIP1 expression in inhibitory neurons via enhancer-mediated mechanisms,with an adjacent peak modulating CCBE1 expression in neuroblasts and granule cells.Transcriptome-wide association studies(TWAS)confirmed PMAIP1's role in the hypothalamus,and Mendelian randomization(MR)validated PMAIP1 and CCBE1 as key brain expression quantitative trait locus(eQTL)effectors.We propose that the variant rs342950505,located within a regulatory peak,modulates PMAIP1 expression in inhibitory neurons,potentially influencing energy homeostasis via hypothalamic regulation.Similarly,CCBE1 may contribute to this process.Conclusions Our results,through systematic dissection of pleiotropic BFT-associated loci,provide a framework to elucidate regulatory mechanisms of complex traits,offering insights into polygenic control through lipid metabolism and neural signaling pathways.
基金supported by the Deanship of Research and Graduate Studies at King Khalid University under Small Research Project grant number RGP1/139/45.
文摘Integrating multiple medical imaging techniques,including Magnetic Resonance Imaging(MRI),Computed Tomography,Positron Emission Tomography(PET),and ultrasound,provides a comprehensive view of the patient health status.Each of these methods contributes unique diagnostic insights,enhancing the overall assessment of patient condition.Nevertheless,the amalgamation of data from multiple modalities presents difficulties due to disparities in resolution,data collection methods,and noise levels.While traditional models like Convolutional Neural Networks(CNNs)excel in single-modality tasks,they struggle to handle multi-modal complexities,lacking the capacity to model global relationships.This research presents a novel approach for examining multi-modal medical imagery using a transformer-based system.The framework employs self-attention and cross-attention mechanisms to synchronize and integrate features across various modalities.Additionally,it shows resilience to variations in noise and image quality,making it adaptable for real-time clinical use.To address the computational hurdles linked to transformer models,particularly in real-time clinical applications in resource-constrained environments,several optimization techniques have been integrated to boost scalability and efficiency.Initially,a streamlined transformer architecture was adopted to minimize the computational load while maintaining model effectiveness.Methods such as model pruning,quantization,and knowledge distillation have been applied to reduce the parameter count and enhance the inference speed.Furthermore,efficient attention mechanisms such as linear or sparse attention were employed to alleviate the substantial memory and processing requirements of traditional self-attention operations.For further deployment optimization,researchers have implemented hardware-aware acceleration strategies,including the use of TensorRT and ONNX-based model compression,to ensure efficient execution on edge devices.These optimizations allow the approach to function effectively in real-time clinical settings,ensuring viability even in environments with limited resources.Future research directions include integrating non-imaging data to facilitate personalized treatment and enhancing computational efficiency for implementation in resource-limited environments.This study highlights the transformative potential of transformer models in multi-modal medical imaging,offering improvements in diagnostic accuracy and patient care outcomes.
基金supported by research grants provided by the National Natural Science Foundation of China(Grant Nos.32101581,32271951,and 32372754)the Hubei Provincial Central Leading Local Special Project(Grant No.2022BGE263)+3 种基金the Key Research and Science and Technology Program of Hubei Province(Grant No.2021BBA098)the Hubei Province Natural Science Foundation(Grant Nos.2023AFB1063 and 2024AFB1057)the Innovation Team Project from Hubei University of Science and Technology(Grant No.2022T02)a PhD grant from the Hubei University of Science and Technology(Grant Nos.BK202002and BK202419).
文摘Osmanthus fragrans Lour.is a well-known aromatic plant widely used as a food ingredient due to its unique floral fragrance and bioactive compounds.To fully utilize O.fragrans resources,we established an O.fragrans multi-omics database called the O.fragrans Information Resource(OfIR:http://yanglab.hzau.edu.cn/OfIR/home/).OfIR is a convenient and comprehensive multi-omics database that efficiently integrates phenotype and genetic variation from 127 O.fragrans cultivars,and provides many easy-to-use analysis tools,including primer design,sequence extraction,multi-sequence alignment,GO and KEGG enrichment analysis,variation annotation,and electronic PCR.Two case studies were used to demonstrate its power to mine candidate genetic variation sites or genes associated with specific traits or regulatory networks.In summary,the multi-omics database OfIR provides a convenient and user-friendly platform for researchers in mining functional genes and contributes to the genetic breeding of O.fragrans.
基金funded by the Chief Scientist Program of Qinghai Province(2024-SF-102)the Joint Special Project of Sanjiangyuan National Park(LHZX-2023-02).
文摘Background As an indigenous livestock species on the Tibetan Plateau,Tibetan sheep exhibit remarkable adaptability to low temperatures and nutrient-scarce environments.During the cold season,Tibetan sheep are typically managed under two feeding regimes:barn feeding(BF)and traditional grazing(TG).However,the molecular mechanisms underlying their adaptation to these distinct management strategies remain unclear.This study aimed to investigate the adaptive strategies of rumen function in Tibetan sheep to cold-season feeding regimes by integrating analyses of rumen morphology,microbiome,metabolome,and transcriptome.Twelve healthy Tibetan sheep with similar body weights were assigned into two groups(BF vs.TG).At the end of the experiment,rumen tissues were subjected to histological observation.Multi-omics techniques were employed to evaluate the effects of cold-season feeding regimes on rumen function in Tibetan sheep.Results The ruminal papilla height,width,and muscular thickness were significantly higher in BF group.The relative abundances of Actinobacteria and Succiniclasticum were significantly elevated in the rumen of BF group,whereas Rikenellaceae,Gracilibacteria,and Lachnospiraceae showed higher abundances in the TG group.Metabolomic analysis identified 19 differential metabolites between the two groups,including upregulated compounds in BF group(fumaric acid,maltose,L-phenylalanine,and L-alanine)and TG group(e.g.,phenylacetic acid,salicyluric acid and ferulic acid).These metabolites were predominantly enriched in phenylalanine metabolism,alanine,aspartate and glutamate metabolism,and phenylalanine,tyrosine and tryptophan biosynthesis pathways.Additionally,210 differentially expressed genes(DEGs)were identified in rumen epithelium:100 upregulated DEGs in the BF group were enriched in nutrient metabolism-related pathways(e.g.,fatty acid degradation and PPAR signaling pathway),while 110 upregulated DEGs in the TG group were associated with immune-related pathways(e.g.,p53 signaling pathway and glutathione metabolism).Conclusions Among these,we observed distinct rumen functional responses to different cold-season feeding regimes in Tibetan sheep and revealed energy allocation strategies mediated by host-microbe interactions.In the BF group,Tibetan sheep adopted a"metabolic efficiency-priority"strategy,driving rumen microbiota to maximize energy capture from high-nutrient diets to support host growth.In contrast,the TG group exhibited an"environmental adaptation-priority"strategy,where rumen microbiota prioritized cellulose degradation and anti-inflammatory functions,reallocating energy toward homeostasis maintenance at the expense of rumen development and growth performance.
文摘BACKGROUND Gastrointestinal(GI)malignancies,including gastric and colorectal cancers,remain one of the primary contributors to cancer-related illness and death globally.Despite the availability of conventional diagnostic tools,early detection and personalized treatment remain significant clinical challenges.Integrated multi-omics methods encompassing genomic,transcriptomic,proteomic,metabolomic,and microbiome profiles have emerged as powerful tools for advancing precision oncology,improving diagnostic accuracy,and informing therapeutic strategies.AIM To investigate the application of multi-omics approaches in the early detection,risk stratification,treatment optimization,and biomarker discovery of GI malignancies.METHODS The systematic review process was conducted in accordance with the PRISMA 2020 guidelines.Five databases,PubMed,ScienceDirect,Scopus,ProQuest,and Web of Science,were searched for studies published in English from 2015 onwards.Eligible studies involved human subjects and focused on multi-omics integration in GI cancers,including biomarker identification,tumor microenvironment analysis,tumor heterogeneity,organoid modeling,and artificial intelligence(AI)-driven analytics.Data extraction included study characteristics,omics modalities,clinical applications,and evaluation of study quality conducted with the Cochrane risk of bias 2.0 instrument.RESULTS A total of 17196 initially identified articles,20 met the inclusion criteria.The findings highlight the superiority of multi-omics platforms over traditional biomarkers(e.g.,carcinoembryonic antigen and carbohydrate antigen 19-9 in detecting early stage GI cancers.Key applications include the identification of circulating tumor DNA,extracellular vesicles,lipidomic and proteomic signatures,and the adoption of AI algorithms to enhance diagnostic precision.Multi-omics analysis has also revealed the mechanisms of immune modulation,tumor microenvironment regulation,metastatic behavior,and drug resistance.Organoid models and microbiota profiling have contributed to personalized therapeutic strategies and immunotherapy optimization.CONCLUSION Multi-omics approaches offer significant advancements in the early diagnosis,prognostic evaluation,and personalized treatment of GI malignancies.Their integration with AI analytics,organoid biobanking,and microbiota modulation provides a pathway for precision oncology research.
基金Supported by 2023 Government-funded Project of the Outstanding Talents Training Program in Clinical Medicine,No.ZF2023165Key Research and Development Projects of Hebei Province,No.18277731DNatural Science Foundation of Hebei Province,No.H202423105.
文摘In this editorial,we discuss the findings reported by Wang et al in the latest issue of the World Journal of Gastrointestinal Oncology.Various research methodologies,including microbiome analysis,assert that the Tzu-Chi Cancer-Antagonizing and Life-Protecting II Decoction of Chinese herbal compounds mitigates inflammatory responses by inhibiting the NF-κB signaling pathway.This action helps maintain the dynamic equilibrium of the intestinal microecology and lessens chemotherapy-induced gastrointestinal damage.The efficacy of these compounds is intimately linked to the composition of intestinal microbes.These compounds regulate intestinal microecology by virtue of their specific compatibility and effectiveness,thereby enhancing the overall therapeutic outcomes of cancer chemotherapy.Nonetheless,the exact mechanisms underlying these effects warrant further investigation.Multi-omics technologies offer a systematic approach to elucidate the mechanisms and effectiveness of Chinese herbal compounds in vivo.This manuscript reviews the application of multi-omics technologies to Chinese herbal compounds and explores their potential role in modulating the gastrointestinal microenvironment following cancer chemotherapy,thus providing a theoretical foundation for their continued use in adjunct cancer treatment.
基金the Central Guidance for Regional Science and Technology Development Projects(YDZJSX2024B010)Research project of Shanxi Provincial Health Commission(2024067)。
文摘Objective Pneumoconiosis,a lung disease caused by irreversible fibrosis,represents a significant public health burden.This study investigates the causal relationships between gut microbiota,gene methylation,gene expression,protein levels,and pneumoconiosis using a multi-omics approach and Mendelian randomization(MR).Methods We analyzed gut microbiota data from MiBioGen and Esteban et al.to assess their potential causal effects on pneumoconiosis subtypes(asbestosis,silicosis,and inorganic pneumoconiosis)using conventional and summary-data-based MR(SMR).Gene methylation and expression data from Genotype-Tissue Expression and eQTLGen,along with protein level data from deCODE and UK Biobank Pharma Proteomics Project,were examined in relation to pneumoconiosis data from FinnGen.To validate our findings,we assessed self-measured gut flora from a pneumoconiosis cohort and performed fine mapping,drug prediction,molecular docking,and Phenome-Wide Association Studies to explore relevant phenotypes of key genes.Results Three core gut microorganisms were identified:Romboutsia(OR=0.249)as a protective factor against silicosis,Pasteurellaceae(OR=3.207)and Haemophilus parainfluenzae(OR=2.343)as risk factors for inorganic pneumoconiosis.Additionally,mapping and quantitative trait loci analyses revealed that the genes VIM,STX8,and MIF were significantly associated with pneumoconiosis risk.Conclusions This multi-omics study highlights the associations between gut microbiota and key genes(VIM,STX8,MIF)with pneumoconiosis,offering insights into potential therapeutic targets and personalized treatment strategies.
基金partially supported by the National Natural Science Foundation of China under Grants 62471493 and 62402257(for conceptualization and investigation)partially supported by the Natural Science Foundation of Shandong Province,China under Grants ZR2023LZH017,ZR2024MF066,and 2023QF025(for formal analysis and validation)+1 种基金partially supported by the Open Foundation of Key Laboratory of Computing Power Network and Information Security,Ministry of Education,Qilu University of Technology(Shandong Academy of Sciences)under Grant 2023ZD010(for methodology and model design)partially supported by the Russian Science Foundation(RSF)Project under Grant 22-71-10095-P(for validation and results verification).
文摘To address the challenge of missing modal information in entity alignment and to mitigate information loss or bias arising frommodal heterogeneity during fusion,while also capturing shared information acrossmodalities,this paper proposes a Multi-modal Pre-synergistic Entity Alignmentmodel based on Cross-modalMutual Information Strategy Optimization(MPSEA).The model first employs independent encoders to process multi-modal features,including text,images,and numerical values.Next,a multi-modal pre-synergistic fusion mechanism integrates graph structural and visual modal features into the textual modality as preparatory information.This pre-fusion strategy enables unified perception of heterogeneous modalities at the model’s initial stage,reducing discrepancies during the fusion process.Finally,using cross-modal deep perception reinforcement learning,the model achieves adaptive multilevel feature fusion between modalities,supporting learningmore effective alignment strategies.Extensive experiments on multiple public datasets show that the MPSEA method achieves gains of up to 7% in Hits@1 and 8.2% in MRR on the FBDB15K dataset,and up to 9.1% in Hits@1 and 7.7% in MRR on the FBYG15K dataset,compared to existing state-of-the-art methods.These results confirm the effectiveness of the proposed model.
文摘Cancer rates are increasing globally,making it more urgent than ever to enhance research and treatment strategies.This study aims to investigate how innovative technology and integrated multi-omics techniques could help improve cancer diagnosis,knowledge,and therapy.A complete literature search was undertaken using PubMed,Elsevier,Google Scholar,ScienceDirect,Embase,and NCBI.This review examined the articles published from 2010 to 2025.Relevant articles were found using keywords and selected using inclusion criteria New sequencing methods,like next-generation sequencing and single-cell analysis,have transformed our ability to study tumor complexity and genetic mutations,paving the way for more precise,personalized treatments.At the same time,imaging technologies such as Positron Emission Tomography(PET)and Magnetic Resonance Imaging(MRI)have made detecting tumors early and tracking treatment progress easier,all while improving patient comfort.Artificial intelligence(AI)and machine learning(ML)are having a significant impact by helping to analyze large volumes of data more efficiently and enhancing diagnostic accuracy.Meanwhile,Clustered Regulatory Interspaced Short Palindromic Repeats(CRISPR/Cas9)gene editing is emerging as a promising tool for directly targeting genes related to cancer,providing new possibilities for treatment.By integrating genomic,transcriptomic,proteomic,and metabolomic data,multi-omics approaches provide researchers with a more comprehensive understanding of the molecular mechanisms driving cancer,thereby facilitating the discovery of novel biomarkers and therapeutic targets.Despite these advancements,additional challenges persist,such as data integration,elevated costs,standardisation concerns,and the intricacies of translating findings into clinical practice,which might prevent wider implementation.Research needs to concentrate on improving these developments and encouraging multidisciplinary cooperation going forward to maximize their possibilities.Personalized cancer therapies will become more successful with ongoing developments,therefore enhancing patient outcomes and quality of life.
