Kernel Bayesian logistic tensor decomposition with automatic rank determination for predicting multiple types of miRNA-disease associations.

Identifying the association and corresponding types of miRNAs and diseases is crucial for studying the molecular mechanisms of disease-related miRNAs. Compared to traditional biological experiments, computational models can not only save time and reduce costs, but also discover potential associations on a large scale. Although some computational models based on tensor decomposition have been proposed, these models usually require manual specification of numerous hyperparameters, leading to a decrease in computational efficiency and generalization ability. Additionally, these linear models struggle to analyze complex, higher-order nonlinear relationships. Based on this, we propose a novel framework, KBLTDARD, to identify potential multiple types of miRNA-disease associations. Firstly, KBLTDARD extracts information from biological networks and high-order association network, and then fuses them to obtain more precise similarities of miRNAs (diseases). Secondly, we combine logistic tensor decomposition and Bayesian methods to achieve automatic hyperparameter search by introducing sparse-induced priors of multiple latent variables, and incorporate auxiliary information to improve prediction capabilities. Finally, an efficient deterministic Bayesian inference algorithm is developed to ensure computational efficiency. Experimental results on two benchmark datasets show that KBLTDARD has better Top-1 precision, Top-1 recall, and Top-1 F1 for new type predictions, and higher AUPR, AUC, and F1 values for new triplet predictions, compared to other state-of-the-art methods. Furthermore, case studies demonstrate the efficiency of KBLTDARD in predicting multiple types of miRNA-disease associations.

Structural Regulation of NiFe LDH under Spontaneous Corrosion to Enhance the Oxygen Evolution Properties.

Electrochemical water splitting holds promise for sustainable hydrogen production but restricted by the sluggish reaction kinetics at the anodic oxygen evolution. Herein, we present a room-temperature spontaneous corrosion strategy to convert inexpensive iron (Fe) on iron foam substrates into highly active and stable self-supporting nickel iron layered hydroxide (NiFe LDH) catalysts. The corrosion evolution mechanisms are elucidated combining ex-situ scanning electron microscopy (SEM) and X-ray photo electron spectroscopy (XPS) techniques, demonstrating precise control over the concentration of Ni2+ and reaction time to achieve controllable micro-structures of NiFe LDH. Taking advantage of the self-supporting morphology and hierarchical micro-/nano- structure, the NiFe LDH with optimized Ni2+ concentration and reaction time exhibits significant small overpotentials of 160 mV and 200 mV for the OER at current densities of 10 mA cm-2 and 100 mA cm-2 respectively, showcasing excellent OER activities. Furthermore, this catalyst demonstrates superior reaction kinetics, high electrochemical stability, and excellent integral water splitting performance when coupled with a commercial Pt/C cathode. The energy-efficient, cost-effective, and scalable spontaneous corrosion strategy opens new avenues for the development of high-electrochemical-interface catalysts.

Optimizing Acute Coronary Syndrome Patient Treatment: Leveraging Gated Transformer Models for Precise Risk Prediction and Management.

BACKGROUND: Acute coronary syndrome (ACS) is a severe cardiovascular disease with globally rising incidence and mortality rates. Traditional risk assessment tools are widely used but are limited due to the complexity of the data. METHODS: This study introduces a gated Transformer model utilizing machine learning to analyze electronic health records (EHRs) for an enhanced prediction of major adverse cardiovascular events (MACEs) in ACS patients. The model's efficacy was evaluated using metrics such as area under the curve (AUC), precision-recall (PR), and F1-scores. Additionally, a patient management platform was developed to facilitate personalized treatment strategies. RESULTS: Incorporating a gating mechanism substantially improved the Transformer model's performance, especially in identifying true-positive cases. The TabTransformer+Gate model demonstrated an AUC of 0.836, a 14% increase in average precision (AP), and a 6.2% enhancement in accuracy, significantly outperforming other deep learning approaches. The patient management platform enabled healthcare professionals to effectively assess patient risks and tailor treatments, improving patient outcomes and quality of life. CONCLUSION: The integration of a gating mechanism within the Transformer model markedly increases the accuracy of MACE risk predictions in ACS patients, optimizes personalized treatment, and presents a novel approach for advancing clinical practice and research.

Kernel Bayesian nonlinear matrix factorization based on variational inference for human-virus protein-protein interaction prediction.

Identification of potential human-virus protein-protein interactions (PPIs) contributes to the understanding of the mechanisms of viral infection and to the development of antiviral drugs. Existing computational models often have more hyperparameters that need to be adjusted manually, which limits their computational efficiency and generalization ability. Based on this, this study proposes a kernel Bayesian logistic matrix decomposition model with automatic rank determination, VKBNMF, for the prediction of human-virus PPIs. VKBNMF introduces auxiliary information into the logistic matrix decomposition and sets the prior probabilities of the latent variables to build a Bayesian framework for automatic parameter search. In addition, we construct the variational inference framework of VKBNMF to ensure the solution efficiency. The experimental results show that for the scenarios of paired PPIs, VKBNMF achieves an average AUPR of 0.9101, 0.9316, 0.8727, and 0.9517 on the four benchmark datasets, respectively, and for the scenarios of new human (viral) proteins, VKBNMF still achieves a higher hit rate. The case study also further demonstrated that VKBNMF can be used as an effective tool for the prediction of human-virus PPIs.

Hierarchical Sea Urchin-like Fe-doped Heazlewoodite for High-Efficient Oxygen Evolution.

Electrochemical water-splitting to produce hydrogen is potential to substitute the traditional industrial coal gasification, but the oxygen evolution kinetics at the anode remains sluggish. In this paper, sea urchin-like Fe doped Ni3S2 catalyst growing on nickel foam (NF) substrate is constructed via a simple two-step strategy, including surface iron activation and post sulfuration process. The NF-Fe-Ni3S2 obtains at temperature of 130 °C (NF-Fe-Ni3S2-130) features nanoneedle-like arrays which are vertically grown on the particles to form sea urchin-like morphology, features high electrochemical surface area. As oxygen evolution catalyst, NF-Fe-Ni3S2-130 exhibits excellent oxygen evolution activities, fast reaction kinetics, and superior reaction stability. The excellent OER performance of sea urchin-like NF-Fe-Ni3S2-130 is mainly ascribed to the high-vertically dispersive of nanoneedles and the existing Fe dopants, which obviously improved the reaction kinetics and the intrinsic catalytic properties. The simple preparation strategy is conducive to establish high-electrochemical-interface catalysts, which shows great potential in renewable energy conversion.

Weighted hypergraph learning and adaptive inductive matrix completion for SARS-CoV-2 drug repositioning.

MOTIVATION: The outbreak of the human coronavirus (SARS-CoV-2) has placed a huge burden on public health and the world economy. Compared with de novo drug discovery, drug repurposing is a promising therapeutic strategy that facilitates rapid clinical treatment decisions, shortens the development process, and reduces costs. RESULTS: In this study, we propose a weighted hypergraph learning and adaptive inductive matrix completion method, WHAIMC, for predicting potential virus-drug associations. Firstly, we integrate multi-source data to describe viruses and drugs from multiple perspectives, including drug chemical structures, drug targets, virus complete genome sequences, and virus-drug associations. Then, WHAIMC establishes an adaptive inductive matrix completion model to improve performance through adaptive learning of similarity relations. Finally, WHAIMC introduces weighted hypergraph learning into adaptive inductive matrix completion to capture higher-order relationships of viruses (or drugs). The results showed that WHAIMC had a strong predictive performance for new virus-drug associations, new viruses, and new drugs. The case study further demonstrates that WHAIMC is highly effective for repositioning antiviral drugs against SARS-CoV-2 and provides a new perspective for virus-drug association prediction. The code and data in this study is freely available at https://github.com/Mayingjun20179/WHAIMC.

Preparation and Characterization of a Novel Tragacanth Gum/Chitosan/Sr-Nano-Hydroxyapatite Composite Membrane.

It is a great challenge to obtain an ideal guided bone regeneration (GBR) membrane. In this study, tragacanth gum (GT) was introduced into a chitosan/nano-hydroxyapatite (CS/n-HA) system. The effects of different component ratios and strontium-doped nano-hydroxyapatite (Sr-HA) on the physical-chemical properties and degradation behavior of the CS/Sr-n-HA/GT ternary composite membrane were investigated using Fourier transform infrared spectroscopy (FT-IR), X-ray diffraction (XRD), scanning electron microscopy (SEM), contact angle, electromechanical universal tester and in vitro soaking in simulated body fluid (SBF). The results showed that CS could be ionically crosslinked with GT through electrostatic interaction, and Sr-n-HA was loaded via hydrogen bond, which endowed the GT/CS/n-HA composite membrane with good tensile strength and hydrophilicity. In addition, the results of immersion in SBF in vitro showed that CS/n-HA/GT composite membranes had different degradation rates and good apatite deposition by investigating the changes in pH value, weight loss, water absorption ratio, SEM morphology observation and tensile strength reduction. All results revealed that the CS/Sr-n-HA/GT (6:2:2) ternary composite membrane possessed the strongest ionic crosslinking of GT and CS, which was expected to obtain more satisfactory GBR membranes, and this study will provide new applications of GT in the field of biomedical membranes.

Cobalt Oxide Arrays Anchored to Copper Foam as Efficient Binder-free Anode for Lithium Ion Batteries.

The development of lithium-ion batteries with simplified assembling steps and fast charge capability is crucial for current battery applications. In this study, we propose a simple in-situ strategy for the construction of high-dispersive cobalt oxide (CoO) nanoneedle arrays, which grow vertically on a copper foam substrate. It is demonstrated that this nanoneedle CoO electrodes provide abundant electrochemical surface area. The resulting CoO arrays directly act as binder-free anodes in lithium-ion batteries with the copper foam functioning as the current collector. The highly-dispersed feature of the nanoneedle arrays enhances the effectiveness of active materials, leading to outstanding rate capability and superior long-term cycling stability. These impressive electrochemical properties are attributed to the highly-dispersed self-standing nanoarrays, the advantages of binder-free constituent, and the high exposed surface area of the copper foam substrate compared to copper foil, which enrich active surface area and facilitate charge transfer. The proposed approach to prepare binder-free lithium-ion battery anodes streamlines the electrode fabrication steps and holds significant promise for the future development of the battery industry.

HONMF: integration analysis of multi-omics microbiome data via matrix factorization and hypergraph.

MOTIVATION: The accumulation of multi-omics microbiome data provides an unprecedented opportunity to understand the diversity of bacterial, fungal, and viral components from different conditions. The changes in the composition of viruses, bacteria, and fungi communities have been associated with environments and critical illness. However, identifying and dissecting the heterogeneity of microbial samples and cross-kingdom interactions remains challenging. RESULTS: We propose HONMF for the integrative analysis of multi-modal microbiome data, including bacterial, fungal, and viral composition profiles. HONMF enables identification of microbial samples and data visualization, and also facilitates downstream analysis, including feature selection and cross-kingdom association analysis between species. HONMF is an unsupervised method based on hypergraph induced orthogonal non-negative matrix factorization, where it assumes that latent variables are specific for each composition profile and integrates the distinct sets of latent variables through graph fusion strategy, which better tackles the distinct characteristics in bacterial, fungal, and viral microbiome. We implemented HONMF on several multi-omics microbiome datasets from different environments and tissues. The experimental results demonstrate the superior performance of HONMF in data visualization and clustering. HONMF also provides rich biological insights by implementing discriminative microbial feature selection and bacterium-fungus-virus association analysis, which improves our understanding of ecological interactions and microbial pathogenesis. AVAILABILITY AND IMPLEMENTATION: The software and datasets are available at https://github.com/chonghua-1983/HONMF.

Study on the Quality of Mixed Silage of Rapeseed with Alfalfa or Myriophyllum.

The objective of this study was to improve the comprehensive rate of utilization of rapeseed (Brassica napus subsp. napus L.), Myriophyllum (Myriophyllum spicatum L.) spicatum and alfalfa (Medicago sativa L.), reduce resource waste and environmental pollution. In this experiment, the effects of different proportions of the mixed silage of rapeseed and alfalfa or M. spicatum on the fermentation and nutritional quality were analyzed and further improved the quality of mixed silage using molasses and urea. Rapeseed was separately silaged with alfalfa and M. spicatum based on the ratios of 3:7, 5:5 and 7:3. After 60 days of mixed silage, the fermentation index and nutrient contents were measured to explore the appropriate ratio of mixed silage. The mixing ratio of rapeseed and alfalfa was better at 3:7: The contents of NH3-N/TN (4.61%), lactic acid (96.46 g·kg-1 dry matter [DM]) were significantly higher (p < 0.05). The crude protein content (118.20 g·kg-1 DM) was the highest (p < 0.05), while the pH (4.56) was the lowest when the mixing ratio of rapeseed and M. spicatum was 7:3. Considering the fermentation and nutrition quality, it is suggested that rapeseed and alfalfa should be mixed as silage at a ratio of 3:7 with 3% molasses and 0.3% urea, and rapeseed and M. spicatum should be mixed as silage at a ratio of 7:3 with 3% molasses.

Effects of initial volatile fatty acid concentrations on process characteristics, microbial communities, and metabolic pathways on solid-state anaerobic digestion.

Solid-state anaerobic digestion (SSAD) is vulnerable to excess volatile fatty acids (VFA), mainly acetate and propionate. The co-effects of VFAs and microbial dynamics under VFA accumulation were investigated in SSAD of pig manure and corn straw. Adding 2 and 4 mg/g acetate or propionate caused initial increases in total VFAs, followed by decreases after day 6, resulting in 'mild' VFA accumulation, while adding 6 mg/g caused similarly increased VFAs, but with no subsequent decrease, causing 'severe' VFA accumulation and poor methanation performance. Mild propionate accumulation promoted acetate consumption, whereas acetate accumulation inhibited propionate degradation by affecting crucial redox reactions. Under severe VFA accumulation, hydrolysis and acidification mainly conducted by acid-tolerant Clostridium sp. exacerbated VFA inhibition, causing a competition between Methanosarcina and Methanosaeta, and impairments of acetoclastic and hydrogenotrophic methanogenesis and interspecies formate transfer. This study provides new insights into mechanisms of VFA accumulation in SSAD, and its effects on methanogenesis.

Logistic tensor decomposition with sparse subspace learning for prediction of multiple disease types of human-virus protein-protein interactions.

Viral infection involves a large number of protein-protein interactions (PPIs) between the virus and the host, and the identification of these PPIs plays an important role in revealing viral infection and pathogenesis. Existing computational models focus on predicting whether human proteins and viral proteins interact, and rarely take into account the types of diseases associated with these interactions. Although there are computational models based on a matrix and tensor decomposition for predicting multi-type biological interaction relationships, these methods cannot effectively model high-order nonlinear relationships of biological entities and are not suitable for integrating multiple features. To this end, we propose a novel computational framework, LTDSSL, to determine human-virus PPIs under different disease types. LTDSSL utilizes logistic functions to model nonlinear associations, sets importance levels to emphasize the importance of observed interactions and utilizes sparse subspace learning of multiple features to improve model performance. Experimental results show that LTDSSL has better predictive performance for both new disease types and new triples than the state-of-the-art methods. In addition, the case study further demonstrates that LTDSSL can effectively predict human-viral PPIs under various disease types.

High-temperature adaptation of an OsNRT2.3 allele is thermoregulated by small RNAs.

Climate change negatively affects crop yield, which hinders efforts to reach agricultural sustainability and food security. Here, we show that a previously unidentified allele of the nitrate transporter gene OsNRT2.3 is required to maintain high yield and high nitrogen use efficiency under high temperatures. We demonstrate that this tolerance to high temperatures in rice accessions harboring the HTNE-2 (high temperature resistant and nitrogen efficient-2) alleles from enhanced translation of the OsNRT2.3b mRNA isoform and the decreased abundance of a unique small RNA (sNRT2.3-1) derived from the 5' untranslated region of OsNRT2.3. sNRT2.3-1 binds to the OsNRT2.3a mRNA in a temperature-dependent manner. Our findings reveal that allelic variation in the 5' untranslated region of OsNRT2.3 leads to an increase in OsNRT2.3b protein levels and higher yield during high-temperature stress. Our results also provide a breeding strategy to produce rice varieties with higher grain yield and lower N fertilizer input suitable for a sustainable agriculture that is resilient against climate change.

Utilizing Cell Culture Assisted Anodization to Fabricate Aluminium Oxide with a Gradient Microstep and Nanopore Structure.

Anodic aluminum oxide (AAO) with a gradient microstep and nanopore structure (GMNP) is fabricated by inversely using cell culture to control the reaction areas in the electrochemical anodization, which shows a larger porosity than that of typical planar AAO. The figure of the microstep is influenced by the cell dehydration temperature which controls the cell shrinkage degree. A GMNP AAO with a diameter of 2.5 cm is achieved. Polymer with a gradient microstep and nanonipple structure is fabricated using the GMNP AAO as the template, which denotes that GMNP AAO could become a broad platform for the structural preparation of various materials with advanced functions.

Prevalence and pathology of Cephalopina titillator infestation in Camelus bactrianus from Xinjiang, China.

BACKGROUND: In camels, nasopharyngeal myiasis is caused by the larvae of Cephalopina titillator, which parasitize the tissues of nasal and paranasal sinuses, pharynx, and larynx. C. titillator infestation adversely affects the health of camels and decreases milk and meat production and even death. However, the C. titillator infestation in Bactrian camels has not been widely studied. METHODS: The present study was conducted to determine the prevalence and risk factors of C. titillator in Bactrian camels of northwestern Xinjiang. Suspected larvae recovered from infested camels were evaluated for C. titillator by microscopy and polymerase chain reaction. Nucleotide sequences of the partial mitochondrial cytochrome c oxidase subunit I (COX1) and cytochrome b (CYTB) genes from the C. titillator of camels were aligned from the NCBI database. Furthermore, the gross and histopathological alterations associated with C. titillator infestation were evaluated via pathological examination. RESULTS: Of 1263 camels examined 685 (54.2%) camels were infested with suspected C. titillator larvae. Different larval stages were topically detected in the nasal passages and pharynx of the camel heads. Microscopy analysis of the pharyngeal mucosa tissue revealed necrotic tissue debris and some inflammatory cells. Molecular detection of the larval COX1 and CYTB genes indicated that pathogen collected in Bactrian camels was C. titillator. The epidemiological study demonstrated that the prevalence rate of C.titillator infestation was significantly higher in camels of Bestierek Town Pasture (67.2%) and Karamagai Town Pasture (63.6%) compared to Kitagel Town Pasture (38.7%) and Qibal Town Pasture (35.8%) (P < 0.05). No significant difference was observed between the prevalence rates in male (52.6%) and female (54.6%) camels (P > 0.05). The prevalence was higher in warm (64.2%) than that in cold (48.4%) seasons (P < 0.001). The prevalence in camels with non-nomadic method (67.2%) was significantly higher than in animals with nomadic method (47.5%) (P < 0.001). The prevalence of C.titillator infestation was significantly higher in animals of aged 5-10 (60.1%) and aged > 10 (61.1%) years old compared to those of aged < 5 (31.7%) years old camels (P < 0.001). CONCLUSION: Our results confirm that there is a high prevalence of C. titillator in Bactrian camels from Xinjiang, closely related to age, season, pasture environment, and husbandry methods. Developing prevention, diagnosis, and control programs to prevent transmission is necessary.

Deciphering discriminative antibiotic resistance genes and pathogens in agricultural soil following chemical and organic fertilizer.

Fertilizers containing rich nutrients can change the profiles of antibiotic resistant pathogens (ARPs) and antibiotic resistance genes (ARGs) in receiving soils; however, the discriminative ARGs and ARPs in agricultural soil following different fertilizer applications remain unknown. Using metagenomic sequencing combined with binning approach, the present study investigated the discriminative ARGs and ARPs under various fertilizer applications (chemical and organic fertilizer) in a 8-year field experiment. VanR, multidrug ARG transporter, vanS, ermA, and arnA were the discriminative ARGs in the chemical fertilizer group, whereas rosB, multidrug transporter, mexW, and aac(3)-I were enhanced in the organic fertilizer group. The metagenomic binning approach revealed that both fertilizer applications caused pathogen proliferation. Chemical fertilizer caused the increase in the pathogenic genus Luteimonas, and organic fertilizer facilitated the proliferation of the pathogenic genera Dokdonella and Pseudomonas. The pathogenic species Pseudomonas_H sp014836765, carrying mexW and multidrug transporter, was enriched only in the organic fertilizer group, indicating that it was a discriminative ARP in the organic fertilizer group. Our results demonstrated that fertilizer application, particularly organic fertilizer application, can facilitate the proliferation of ARGs and ARPs in the receiving soil, posing the risk of the development and spread of soil-borne ARPs.

Hypergraph clustering based on modularity feature projection for high-order relationship community detection of microorganisms.

Microbial community is an important part of organisms or ecosystems to maintain health and stability. Analyzing the interaction of microorganisms in the ecosystem and mining the co-occurrence module of the microbial community can deepen the understanding of microbial community function. This could also improve the ability to manipulate the microbial community, thus provide new means for ecological restoration, disease treatment and drug development. Instead of the investigations of pairwise relationships, more and more studies have realized that the higher-order interactions may play important roles in explaining the diversity and complexity of the community. In this study, a hypergraph clustering (HCMFP) based on modularity feature projection is proposed to detect the microbial community in higher-order interaction network among microbes. Specifically, HCMFP uses information entropy to mine the higher-order logical relationships among microbes, and constructs a hypergraph learning model based on modularity feature projection to detect the microbial community. The experimental results show that compared with other methods, HCMFP has better clustering performance and reliable convergence speed. The proposed method is an effective tool for high-order organizations in microbial interaction network. The code and data in this study is freely available at https://github.com/Mayingjun20179/ HCMFP.

Generalized matrix factorization based on weighted hypergraph learning for microbe-drug association prediction.

The complex and diverse microbial communities are closely related to human health, and the research of microbial communities plays an increasingly critical role in drug development and precision medicine. Identifying potential microbe-drug associations not only benefits drug discovery and clinical therapy, but also contributes to a better understanding of the mechanisms of action of microbes. Compared with the complexity and high cost of biological experiments, computational methods can quickly and efficiently predict potential microbe-drug associations, which could be a useful complement to experimental methods. In this study, we propose a generalized matrix factorization based on weighted hypergraph learning, WHGMF, to predict potential microbial-drug associations. First, we integrate multi-omics data to compute multiple features of microbes and drugs, including functional and semantic similarity of microbes, structural similarity of drugs, and microbe-drug association information. Second, the hypergraph is constructed by using strong neighborhood information, and to improve the performance of the hypergraph, the simple volume is adopted to calculate the hyperedge weight. Finally, hypergraph regularization is introduced for the generalized matrix factorization model, and high-order structural information is used to improve the representation ability of low-dimensional features. Results from multiple experiments demonstrate that WHGMF not only accurately predicts potential microbe-drug associations, but also has considerable adaptability to class-imbalanced datasets. In addition, WHGMF is also suitable for the prediction of new drugs and new microbes. A case study further demonstrates the effectiveness of our method. The code and data in this study are freely available at https://github.com/Mayingjun20179/WHGMF.

DeepMNE: Deep Multi-Network Embedding for lncRNA-Disease Association Prediction.

Long non-coding RNA (lncRNA) participates in various biological processes, hence its mutations and disorders play an important role in the pathogenesis of multiple human diseases. Identifying disease-related lncRNAs is crucial for the diagnosis, prevention, and treatment of diseases. Although a large number of computational approaches have been developed, effectively integrating multi-omics data and accurately predicting potential lncRNA-disease associations remains a challenge, especially regarding new lncRNAs and new diseases. In this work, we propose a new method with deep multi-network embedding, called DeepMNE, to discover potential lncRNA-disease associations, especially for novel diseases and lncRNAs. DeepMNE extracts multi-omics data to describe diseases and lncRNAs, and proposes a network fusion method based on deep learning to integrate multi-source information. Moreover, DeepMNE complements the sparse association network and uses kernel neighborhood similarity to construct disease similarity and lncRNA similarity networks. Furthermore, a graph embedding method is adopted to predict potential associations. Experimental results demonstrate that compared to other state-of-the-art methods, DeepMNE has a higher predictive performance on new associations, new lncRNAs and new diseases. Besides, DeepMNE also elicits a considerable predictive performance on perturbed datasets. Additionally, the results of two different types of case studies indicate that DeepMNE can be used as an effective tool for disease-related lncRNA prediction. The code of DeepMNE is freely available at https://github.com/Mayingjun20179/ DeepMNE.

Hypergraph-based logistic matrix factorization for metabolite-disease interaction prediction.

MOTIVATION: Function-related metabolites, the terminal products of the cell regulation, show a close association with complex diseases. The identification of disease-related metabolites is critical to the diagnosis, prevention and treatment of diseases. However, most existing computational approaches build networks by calculating pairwise relationships, which is inappropriate for mining higher-order relationships. RESULTS: In this study, we presented a novel approach with hypergraph-based logistic matrix factorization, HGLMF, to predict the potential interactions between metabolites and disease. First, the molecular structures and gene associations of metabolites and the hierarchical structures and GO functional annotations of diseases were extracted to build various similarity measures of metabolites and diseases. Next, the kernel neighborhood similarity of metabolites (or diseases) was calculated according to the completed interactive network. Second, multiple networks of metabolites and diseases were fused, respectively, and the hypergraph structures of metabolites and diseases were built. Finally, a logistic matrix factorization based on hypergraph was proposed to predict potential metabolite-disease interactions. In computational experiments, HGLMF accurately predicted the metabolite-disease interaction, and performed better than other state-of-the-art methods. Moreover, HGLMF could be used to predict new metabolites (or diseases). As suggested from the case studies, the proposed method could discover novel disease-related metabolites, which has been confirmed in existing studies. AVAILABILITY AND IMPLEMENTATION: The codes and dataset are available at: https://github.com/Mayingjun20179/HGLMF. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.