RÉSUMÉ
BACKGROUND: Bacillus cereus (B. cereus) is a widespread conditional pathogen that affects food safety and human health. Conventional methods such as bacteria culture and polymerase chain reaction (PCR) are difficult to use for rapid identification of bacterial spores because of the relatively long analysis times. From a human health perspective, there is an urgent need to develop an ultrasensitive, rapid, and accurate method for the detection of B. cereus spores. RESULTS: The study proposed a new method for rapidly and sensitively detecting the biomarkers of bacterial spores via surface-enhanced Raman spectroscopy (SERS) combined with electrochemical enrichment. The 2,6-Pyridinedicarboxylic acid (DPA) was used as the model analyte to acts as a biomarker of B. cereus spores. The SERS substrate was developed via the in-situ generation of Ag nanoparticles (AgNPs) in a cuttlebone-derived organic matrix (CDOM). Because of the depletion of chitin reduction sites on the CDOM, the pores of the porous channels expanded. The pores diameter of the AgNPs/CDOM porous channel was found to be in the range of 0.7-1.3 nm through molecular diffusion experiments. Based on the porosity of AgNPs/CDOM substrates and the high sensitivity of SERS substrates, the sensor can rapidly and accurately electronically enrich DPA in 40 s with the limit of detection (LOD) of 0.3 nM. SIGNIFICANCE: The results demonstrate that electrochemically assisted SERS substrates can be served as a high sensitivity electrochemical-enrichment device for the rapid and sensitive detection of bacterial spores with minimal interference from potentially coexisting species in biological samples. In this study, it opens up a platform to explore the application of porous channels in natural bio-derived materials in the field of food safety.
Sujet(s)
Bacillus cereus , Marqueurs biologiques , Argent , Analyse spectrale Raman , Spores bactériens , Bacillus cereus/isolement et purification , Bacillus cereus/métabolisme , Analyse spectrale Raman/méthodes , Spores bactériens/isolement et purification , Spores bactériens/composition chimique , Argent/composition chimique , Porosité , Marqueurs biologiques/analyse , Nanoparticules métalliques/composition chimique , Acides picoliniques/analyse , Acides picoliniques/composition chimique , Limite de détection , Propriétés de surfaceRÉSUMÉ
Hyaline cartilage fibrosis is typically considered an end-stage pathology of osteoarthritis (OA), which results in changes to the extracellular matrix. However, the mechanism behind this is largely unclear. Here, we found that the RNA helicase DDX5 was dramatically downregulated during the progression of OA. DDX5 deficiency increased fibrosis phenotype by upregulating COL1 expression and downregulating COL2 expression. In addition, loss of DDX5 aggravated cartilage degradation by inducing the production of cartilage-degrading enzymes. Chondrocyte-specific deletion of Ddx5 led to more severe cartilage lesions in the mouse OA model. Mechanistically, weakened DDX5 resulted in abundance of the Fn1-AS-WT and Plod2-AS-WT transcripts, which promoted expression of fibrosis-related genes (Col1, Acta2) and extracellular matrix degradation genes (Mmp13, Nos2 and so on), respectively. Additionally, loss of DDX5 prevented the unfolding Col2 promoter G-quadruplex, thereby reducing COL2 production. Together, our data suggest that strategies aimed at the upregulation of DDX5 hold significant potential for the treatment of cartilage fibrosis and degradation in OA.
Sujet(s)
Épissage alternatif , DEAD-box RNA helicases , Fibrose , G-quadruplexes , Arthrose , Animaux , DEAD-box RNA helicases/génétique , DEAD-box RNA helicases/métabolisme , Souris , Arthrose/anatomopathologie , Arthrose/génétique , Arthrose/métabolisme , Fibrose/métabolisme , Fibrose/génétique , Fibrose/anatomopathologie , Humains , Cartilage articulaire/anatomopathologie , Cartilage articulaire/métabolisme , Chondrocytes/métabolisme , Chondrocytes/anatomopathologie , Modèles animaux de maladie humaine , MâleRÉSUMÉ
Although biologics have achieved tremendous success in the treatment of psoriasis and revolutionized the clinical management of the disease, certain issues arise during treatments, including the phenotypic switch from psoriasis to other skin disorders and the recurrence of psoriasis after the cessation of biologic treatment. Here we provide a concise overview of the roles of keratinocytes in the pathogenesis of psoriasis, elucidate the involvement of keratinocytes in the phenotypic switch and relapse of psoriasis, and address the challenges encountered in both basic and clinical research on psoriasis.
Sujet(s)
Kératinocytes , Phénotype , Psoriasis , Récidive , Psoriasis/immunologie , Psoriasis/anatomopathologie , Humains , Kératinocytes/immunologie , AnimauxRÉSUMÉ
Skin injury always results in fibrotic, non-functional scars in adults. Although multiple factors are well-known contributors to scar formation, the precise underlying mechanisms remain elusive. This review aims to elucidate the intricacies of the wound healing process, summarize the known factors driving skin cells in wounds toward a scarring fate, and particularly to discuss the impact of fibroblast heterogeneity on scar formation. To the end, we explore potential therapeutic interventions used in the treatment of scarring wounds.
Sujet(s)
Cicatrice , Peau , Adulte , Humains , Cicatrice/thérapie , Cicatrice/anatomopathologie , Peau/anatomopathologie , Cicatrisation de plaie , Fibroblastes/anatomopathologieRÉSUMÉ
Multiview spectral clustering, renowned for its spatial learning capability, has garnered significant attention in the data mining field. However, existing methods assume that the optimal consensus adjacency matrix is confined within the space spanned by each view's adjacency matrix. This constraint restricts the feasible domain of the algorithm and hinders the exploration of the optimal consensus adjacency matrix. To address this limitation, we propose a novel and convex strategy, termed the consensus neighbor strategy, for learning the optimal consensus adjacency matrix. This approach constructs the optimal consensus adjacency matrix by capturing the consensus local structure of each sample across all views, thereby expanding the search space and facilitating the discovery of the optimal consensus adjacency matrix. Furthermore, we introduce the concept of a correlation measuring matrix to prevent trivial solution. We develop an efficient iterative algorithm to solve the resulting optimization problem, benefitting from the convex nature of our model, which ensures convergence to a global optimum. Experimental results on 16 multiview datasets demonstrate that our proposed algorithm surpasses state-of-the-art methods in terms of its robust consensus representation learning capability. The code of this article is uploaded to https://github.com/PhdJiayiTang/Consensus-Neighbor-Strategy.git.
RÉSUMÉ
Deep unrolling networks (DUNs) have emerged as a promising approach for solving compressed sensing (CS) problems due to their superior explainability, speed, and performance compared to classical deep network models. However, the CS performance in terms of efficiency and accuracy remains a principal challenge for approaching further improvements. In this paper, we propose a novel deep unrolling model, SALSA-Net, to solve the image CS problem. The network architecture of SALSA-Net is inspired by unrolling and truncating the split augmented Lagrangian shrinkage algorithm (SALSA) which is used to solve sparsity-induced CS reconstruction problems. SALSA-Net inherits the interpretability of the SALSA algorithm while incorporating the learning ability and fast reconstruction speed of deep neural networks. By converting the SALSA algorithm into a deep network structure, SALSA-Net consists of a gradient update module, a threshold denoising module, and an auxiliary update module. All parameters, including the shrinkage thresholds and gradient steps, are optimized through end-to-end learning and are subject to forward constraints to ensure faster convergence. Furthermore, we introduce learned sampling to replace traditional sampling methods so that the sampling matrix can better preserve the feature information of the original signal and improve sampling efficiency. Experimental results demonstrate that SALSA-Net achieves significant reconstruction performance compared to state-of-the-art methods while inheriting the advantages of explainable recovery and high speed from the DUNs paradigm.
Sujet(s)
Algorithmes , Imagerie par résonance magnétique , Imagerie par résonance magnétique/méthodes , , Traitement d'image par ordinateur/méthodesRÉSUMÉ
How pattern recognition receptors NOD1 and NOD2 sense bacterial muropeptides from extracellular bacteria to drive keratinocyte inflammation remains unclear. In this issue of Immunity, Bharadwaj et al. show that the solute carrier 46A2 (SLC46A2) delivers DAP-muropeptides into the cytosol to drive NOD1 activation in keratinocytes and elicit skin inflammation during psoriasis.
Sujet(s)
Inflammation , Récepteurs de reconnaissance de motifs moléculaires , Humains , Protéine adaptatrice de signalisation NOD1/métabolisme , Protéine adaptatrice de signalisation NOD2/métabolismeRÉSUMÉ
The finite inverted beta mixture model (IBMM) has been proven to be efficient in modeling positive vectors. Under the traditional variational inference framework, the critical challenge in Bayesian estimation of the IBMM is that the computational cost of performing inference with large datasets is prohibitively expensive, which often limits the use of Bayesian approaches to small datasets. An efficient alternative provided by the recently proposed stochastic variational inference (SVI) framework allows for efficient inference on large datasets. Nevertheless, when using the SVI framework to address the non-Gaussian statistical models, the evidence lower bound (ELBO) cannot be explicitly calculated due to the intractable moment computation. Therefore, the algorithm under the SVI framework cannot directly use stochastic optimization to optimize the ELBO, and an analytically tractable solution cannot be derived. To address this problem, we propose an extended version of the SVI framework with more flexibility, namely, the extended SVI (ESVI) framework. This framework can be used in many non-Gaussian statistical models. First, some approximation strategies are applied to further lower the ELBO to avoid intractable moment calculations. Then, stochastic optimization with noisy natural gradients is used to optimize the lower bound. The excellent performance and effectiveness of the proposed method are verified in real data evaluation.
RÉSUMÉ
Aberrant RNA splicing in keratinocytes drives inflammatory skin disorders. In the present study, we found that the RNA helicase DDX5 was downregulated in keratinocytes from the inflammatory skin lesions in patients with atopic dermatitis and psoriasis, and that mice with keratinocyte-specific deletion of Ddx5 (Ddx5∆KC) were more susceptible to cutaneous inflammation. Inhibition of DDX5 expression in keratinocytes was induced by the cytokine interleukin (IL)-17D through activation of the CD93-p38 MAPK-AKT-SMAD2/3 signaling pathway and led to pre-messenger RNA splicing events that favored the production of membrane-bound, intact IL-36 receptor (IL-36R) at the expense of soluble IL-36R (sIL-36R) and to the selective amplification of IL-36R-mediated inflammatory responses and cutaneous inflammation. Restoration of sIL-36R in Ddx5∆KC mice with experimental atopic dermatitis or psoriasis suppressed skin inflammation and alleviated the disease phenotypes. These findings indicate that IL-17D modulation of DDX5 expression controls inflammation in keratinocytes during inflammatory skin diseases.
Sujet(s)
Eczéma atopique , Interleukine-27 , Psoriasis , Souris , Animaux , Interleukine-27/métabolisme , Eczéma atopique/génétique , Eczéma atopique/anatomopathologie , Kératinocytes/métabolisme , Peau/anatomopathologie , Psoriasis/génétique , Psoriasis/anatomopathologie , Inflammation/métabolismeRÉSUMÉ
In order to closely fit the characteristics of continuing education, the development of continuing education teaching activities under the network background should not only be combined with the characteristics of professional adult education but also make reasonable use of modern teaching models in the actual teaching process. Based on the community detection algorithm in complex networks, this article makes thorough research and analysis on the complexity of Chinese continuing education by using complex network technology. By establishing the characteristics of vertex degree distribution, average path length, and clustering coefficient of complex networks, it is confirmed that Chinese continuing education has scale-free network characteristics and small-world network characteristics. The three aspects of relationship strength comprehensively analyze the information dissemination speed, scope, interpretation, and application; through the combination of the ant colony algorithm and complex network technology, multiple information dissemination paths are abstracted in Chinese continuing education. The research shows that the application of complex network algorithms can effectively improve the speed and quality of continuing education in China. It is found that the government should increase the number of adult education projects and improve the level of project categories, form key adult education research basis to promote the diversification of research subjects, expand the space for adult education projects to balance regional and provincial differences and attach importance to basic research on adult education, and integrate applied research.
Sujet(s)
Algorithmes , Formation continue , Adulte , Chine , Analyse de regroupements , HumainsRÉSUMÉ
Over the past decades, tremendous success in the treatment of psoriasis has been achieved using biologics, such as neutralizing antibodies against TNF/TNFR, IL-23, and IL-17A/IL-17RA. Although psoriatic skin lesions appear to resolve after treatment with these biologics, lesions often recur after therapy is discontinued or during therapy. Memory T cells residing in the skin have been considered as the major driver of psoriasis relapse. However, whether structural cells in the skin such as keratinocytes and fibroblasts are involved in the relapse of psoriasis is unknown. In this review, we outline the therapeutic rationale of biologics used in the treatment of psoriasis, summarize different clinical features of psoriasis relapse on the basis of preclinical and clinical data, and specifically discuss how memory T cells and structural cells in the skin are involved in psoriasis relapse. Finally, we discuss the future challenges in the basic or clinical research on psoriasis.
RÉSUMÉ
Psoriasis is a complex chronic inflammatory skin disease with unclear molecular mechanisms. We found that the Src homology-2 domain-containing protein tyrosine phosphatase-2 (SHP2) was highly expressed in both psoriatic patients and imiquimod (IMQ)-induced psoriasis-like mice. Also, the SHP2 allosteric inhibitor SHP099 reduced pro-inflammatory cytokine expression in PBMCs taken from psoriatic patients. Consistently, SHP099 significantly ameliorated IMQ-triggered skin inflammation in mice. Single-cell RNA sequencing of murine skin demonstrated that SHP2 inhibition impaired skin inflammation in myeloid cells, especially macrophages. Furthermore, IMQ-induced psoriasis-like skin inflammation was significantly alleviated in myeloid cells (monocytes, mature macrophages, and granulocytes)-but not dendritic cells conditional SHP2 knockout mice. Mechanistically, SHP2 promoted the trafficking of toll-like receptor 7 (TLR7) from the Golgi to the endosome in macrophages by dephosphorylating TLR7 at Tyr1024, boosting the ubiquitination of TLR7 and NF-κB-mediated skin inflammation. Importantly, Tlr7 point-mutant knock-in mice showed an attenuated psoriasis-like phenotype compared to wild-type littermates following IMQ treatment. Collectively, our findings identify SHP2 as a novel regulator of psoriasis and suggest that SHP2 inhibition may be a promising therapeutic approach for psoriatic patients.
Sujet(s)
Psoriasis , Récepteur de type Toll-7 , Animaux , Modèles animaux de maladie humaine , Humains , Imiquimod/effets indésirables , Imiquimod/métabolisme , Souris , Peau , Récepteur de type Toll-7/métabolisme , Récepteur de type Toll-7/usage thérapeutiqueRÉSUMÉ
A Bayesian nonparametric approach for estimation of a Dirichlet process (DP) mixture of generalized inverted Dirichlet distributions [i.e., an infinite generalized inverted Dirichlet mixture model (InGIDMM)] has been proposed. The generalized inverted Dirichlet distribution has been proven to be efficient in modeling the vectors that contain only positive elements. Under the classical variational inference (VI) framework, the key challenge in the Bayesian estimation of InGIDMM is that the expectation of the joint distribution of data and variables cannot be explicitly calculated. Therefore, numerical methods are usually applied to simulate the optimal posterior distributions. With the recently proposed extended VI (EVI) framework, we introduce lower bound approximations to the original variational objective function in the VI framework such that an analytically tractable solution can be derived. Hence, the problem in numerical simulation has been overcome. By applying the DP mixture technique, an InGIDMM can automatically determine the number of mixture components from the observed data. Moreover, the DP mixture model with an infinite number of mixture components also avoids the problems of underfitting and overfitting. The performance of the proposed approach is demonstrated with both synthesized data and real-life data applications.
RÉSUMÉ
Gasdermin-D (GSDMD), the executioner of pyroptotic cell death when it is cleaved by inflammatory caspases, plays a crucial role in host defense and the response to danger signals. So far, there are no known mechanisms, other than cleavage, for regulating GSDMD. Here, we show that tripartite motif protein TRIM21 acts as a positive regulator of GSDMD-dependent pyroptosis. TRIM21 interacted with GSDMD via its PRY-SPRY domain, maintaining GSDMD stable expression in resting cells yet inducing the N-terminus of GSDMD (GSDMD-N) aggregation during pyroptosis. TRIM21-deficient cells displayed a reduced cell death in response to NLRP3 or NLRC4 inflammasome activation. Genetic ablation of TRIM21 in mice conferred protection from LPS-induced inflammation and dextran sulfate sodium-induced colitis. Therefore, TRIM21 plays an essential role in GSDMD-mediated pyroptosis and may be a viable target for controlling and treating inflammation-associated diseases.
Sujet(s)
Protéines et peptides de signalisation intracellulaire , Pyroptose , Animaux , Inflammasomes/métabolisme , Inflammation , Protéines et peptides de signalisation intracellulaire/génétique , Protéines et peptides de signalisation intracellulaire/métabolisme , Souris , Protéines de liaison aux phosphates/génétique , Protéines de liaison aux phosphates/métabolismeRÉSUMÉ
Psoriasis is a common chronic inflammatory skin disease characterized by abnormal proliferation/differentiation of keratinocytes and excessive immune cell infiltration in the dermis and epidermis. Over the past 2 decades, immune cells have been considered as the main driver of psoriasis because the neutralizing antibodies targeting the IL-23/IL-17 axis that regulates cross-talk between dendritic cells and T cells achieve tremendous success in the treatment of psoriasis. However, whether keratinocyte would be a driver of psoriasis or just an executor in response to immune cells is still under debate. In this review, we focus on the recent advances in the identification of keratinocyte as a trigger of psoriasis, summarize on the role of keratinocytes in self-perpetuating loop to maintain inflammation in psoriasis, and then discuss the possible roles of keratinocytes in the relapse of psoriasis.
Sujet(s)
Prédisposition aux maladies , Kératinocytes/métabolisme , Psoriasis/étiologie , Psoriasis/métabolisme , Animaux , Cytokines/métabolisme , Prédisposition aux maladies/immunologie , Humains , Médiateurs de l'inflammation/métabolisme , Kératinocytes/immunologie , Psoriasis/anatomopathologie , Récidive , Transduction du signal , Lymphocytes T/immunologie , Lymphocytes T/métabolismeRÉSUMÉ
For most of the non-Gaussian statistical models, the data being modeled represent strongly structured properties, such as scalar data with bounded support (e.g., beta distribution), vector data with unit length (e.g., Dirichlet distribution), and vector data with positive elements (e.g., generalized inverted Dirichlet distribution). In practical implementations of non-Gaussian statistical models, it is infeasible to find an analytically tractable solution to estimating the posterior distributions of the parameters. Variational inference (VI) is a widely used framework in Bayesian estimation. Recently, an improved framework, namely, the extended VI (EVI), has been introduced and applied successfully to a number of non-Gaussian statistical models. EVI derives analytically tractable solutions by introducing lower bound approximations to the variational objective function. In this paper, we compare two approximation strategies, namely, the multiple lower bounds (MLBs) approximation and the single lower bound (SLB) approximation, which can be applied to carry out the EVI. For implementation, two different conditions, the weak and the strong conditions, are discussed. Convergence of the EVI depends on the selection of the lower bound, regardless of the choice of weak or strong condition. We also discuss the convergence properties to clarify the differences between MLB and SLB. Extensive comparisons are made based on some EVI-based non-Gaussian statistical models. Theoretical analysis is conducted to demonstrate the differences between the weak and strong conditions. Experimental results based on real data show advantages of the SLB approximation over the MLB approximation.
RÉSUMÉ
T helper 17 (TH17) cells and interleukin-17A (IL-17A) produced by them are critical in autoinflammatory diseases, such as psoriasis. IL-17A has been shown to signal through IL-17 receptor A/IL-17 receptor C (IL-17RA/IL-17RC) complex to drive inflammatory responses. However, in a psoriasis model, we found that Il17rc deficiency did not completely ameliorate the disease, suggesting another receptor. In search for another IL-17A-interacting receptor, we found that IL-17RD directly bound IL-17A but not IL-17F or IL-17A/F heterodimer and formed a heterodimer with IL-17RA. IL-17A-, but not IL-17F- or IL-17A/F-, mediated gene expression was defective in Il17rd-deficient keratinocytes. Il17rd deficiency in nonhemopoietic cells attenuated imiquimod-induced psoriasis-like skin inflammation. Although IL-17RC and IL-17RD differentially activated IL-17A-dependent signaling and gene expression, their compound mutation led to complete deficits in keratinocytes. IL-23 was found induced by IL-17A in keratinocytes, dependent on both IL-17RC and IL-17RD, suggesting feed-forward regulation of IL-23/IL-17 axis in psoriasis. Together, IL-17RD constitutes a second functional receptor for IL-17A and, together with IL-17RC, mediates the proinflammatory gene expression downstream of IL-17A.
Sujet(s)
Interleukine-17/immunologie , Psoriasis/immunologie , Récepteurs aux interleukines/immunologie , Animaux , Cellules cultivées , Femelle , Cellules HEK293 , Humains , Imiquimod , Kératinocytes/immunologie , Souris de lignée C57BL , Souris knockout , Psoriasis/induit chimiquement , Récepteurs aux interleukines/génétique , Peau/immunologieRÉSUMÉ
Aggregation behavior of proteins on the surface of gold nanoparticles (AuNPs) has been extensively studied for its promising applications in biosensing, bioimaging, photodynamic therapy, drug delivery, etc. In this work, we studied adsorption kinetics of an antimicrobial protein, regenerating islet-derived protein 3-alpha (REG3A), on the surface of as-synthesized citrate-capped AuNPs under the influence of lipopolysaccharides (LPSs), with a combined method of UV-vis spectroscopy, multivariate analysis, and molecular dockings. In the AuNPs-REG3A binary system, a component with an "up-and-down" signal was detected by the in-depth data analysis on time-resolved spectroscopic data, corresponding to the protein agglomeration and exfoliation observed in transmission electron microscopy and atomic force microscopy experiments. Intriguingly, LPSs can rescue the spectral oddity-the adsorption pattern in the AuNPs-REG3A-LPS ternary system becomes normal and similar to a typical single-layer mode as in our previous study of the serum albumin-AuNP system ( Ren , X. ; et al., Spectrosc. Lett. , 2016 , 49 , 434 - 443 ). The following molecular modeling suggests that LPS molecules mainly interact with three segments of REG3A amino acid sequences, i.e., P109-T110-Q111-G112, P115-N116, and P137-S138-T139. The latter two protein-ligand interactions impair the REG3A-REG3A protein-protein interaction between the two subunits (E114-P115-N116-G117-E118 and N136-P137-S138-T139-I140). Thus, our results elucidate the LPS inhibitory effect on fibrous protein self-aggregation at the AuNP surface, and molecular dockings give a plausible mechanism to rationalize the competition among protein-protein and protein-ligand interactions.
Sujet(s)
Lipopolysaccharides/composition chimique , Nanoparticules métalliques/composition chimique , Protéines associées à la pancréatite/métabolisme , Multimérisation de protéines/effets des médicaments et des substances chimiques , Adsorption , Or/composition chimique , Humains , Simulation de docking moléculaire , Analyse multifactorielle , Agrégats de protéines , Résonance plasmonique de surfaceRÉSUMÉ
The appropriate inflammatory response is essential for normal wound repair, and skin commensal Staphylococcus epidermidis has been shown to regulate TLR3-mediated inflammatory response to maintain skin homeostasis after injury. However, the underlying mechanism by which S. epidermidis regulates wound-induced inflammation remains largely unexplored. In this study we identified a previously unknown lipopeptide 78 (LP78) from S. epidermidis and showed that LP78 inhibited TLR3-mediated skin inflammation to promote wound healing. Skin injury activated TLR3/NF-κB to promote the interaction of p65 and PPARγ in nuclei and then initiated the inflammatory response in keratinocytes. LP78 activated TLR2-SRC to induce ß-catenin phosphorylation at Tyr654 The phospho-ß-catenin translocated into nuclei to bind to PPARγ, thus disrupting the interaction between p65 and PPARγ. The disassociation between p65 and PPARγ reduced the expression of TLR3-induced inflammatory cytokines in skin wounds of normal and diabetic mice, which correlated with accelerated wound healing. Our data demonstrate that S. epidermidis-derived LP78 inhibits skin inflammation to promote wound healing and suggest that LP78 might be a potential compound for the treatment of delayed or unhealed wounds.
Sujet(s)
Inflammation/traitement médicamenteux , Lipopeptides/pharmacologie , Peau/effets des médicaments et des substances chimiques , Staphylococcus epidermidis/composition chimique , bêta-Caténine/métabolisme , Animaux , Cellules cultivées , Humains , Inflammation/métabolisme , Souris , Souris de lignée C57BL , Souris knockout , Phosphorylation/effets des médicaments et des substances chimiques , Peau/métabolisme , Cicatrisation de plaie/effets des médicaments et des substances chimiquesRÉSUMÉ
In this paper, we develop a novel variational Bayesian learning method for the Dirichlet process (DP) mixture of the inverted Dirichlet distributions, which has been shown to be very flexible for modeling vectors with positive elements. The recently proposed extended variational inference (EVI) framework is adopted to derive an analytically tractable solution. The convergency of the proposed algorithm is theoretically guaranteed by introducing single lower bound approximation to the original objective function in the EVI framework. In principle, the proposed model can be viewed as an infinite inverted Dirichlet mixture model that allows the automatic determination of the number of mixture components from data. Therefore, the problem of predetermining the optimal number of mixing components has been overcome. Moreover, the problems of overfitting and underfitting are avoided by the Bayesian estimation approach. Compared with several recently proposed DP-related methods and conventional applied methods, the good performance and effectiveness of the proposed method have been demonstrated with both synthesized data and real data evaluations.