Local Similarity Research Articles

Quality control of eddy covariance fluxes of two ecosystem types with local flux-variance similarity functions in West Africa

Ecosystem capacity studies for carbon sequestration and water extraction are largely conditioned by Eddy Covariance (EC) measurement availability and quality. Our study aims at evaluating several procedures of EC data qualification. Indeed, quality control of EC data is generally ensured by stationarity and well-developed turbulence tests. The latter is based on integral turbulence characteristic (ITC) models. In this study, we revisited the quality control procedure using locally established ITC models for wind speed components and atmospheric scalars (temperature, humidity and carbon dioxide molar densities) in unstable and stable conditions. We also considered the interdependence of flux qualities because of the corrections applied to fluxes during their preprocessing. The analyses were carried out using one year of data collected above a woodland forest and mixed crop sites, both situated in Northern Benin, West Africa. From these analyzes we note that the quality flags vary depending on the use of local or existing ITC models. Also, the test of turbulence development on atmospheric scalars, previously neglected, is of great importance in the quality control of data. Thus, taking into account the ITC models associated with the atmospheric scalars established locally in the well-developed turbulence test as well as taking into account the interdependence of the flux quality have considerably improved the standard quality control procedure used in effectively filtering unrealistic flux data while keeping a considerable percentage of good and medium quality data.

The application of improved DTW algorithm in sports posture recognition

Sports posture recognition plays a crucial role in modern sports science and training. Posture recognition and analysis plays a positive role in improving sports quality and ensuring sports safety. However, existing recognition technologies still have poor recognition and accuracy in large amounts of posture data. Therefore, to further improve the performance of the existing posture recognition techniques, this study assumes that postures during movement can be effectively represented through the time series of skeletal key points, and the local similarity of these postures can be captured through the Dynamic Time Warping (DTW) algorithm. Based on this assumption, the existing DTW algorithm is improved by introducing the K-Nearest Neighbor (KNN) algorithm and combining it with Principal Component Analysis (PCA) for feature dimensionality reduction. A novel algorithmic model for postures recognition is proposed. The experimental results showed that the improved algorithm performed well in postures recognition rate and accuracy. Especially, when the k value was 5, the recognition rate reached up to 89%, and the accuracy reached 87%. Compared with the existing algorithm, the improved KNN-DTW algorithm has significant improvement in accuracy and computational efficiency. In summary, the new algorithm shows significant advantages in terms of accuracy and stability, providing a powerful tool for the analysis of athletic postures in the field of sports. Meanwhile, this research result has important application prospects in fields such as sports training, sports medicine, and virtual reality.

RNAtango: Analysing and comparing RNA 3D structures via torsional angles.

RNA molecules, essential for viruses and living organisms, derive their pivotal functions from intricate 3D structures. To understand these structures, one can analyze torsion and pseudo-torsion angles, which describe rotations around bonds, whether real or virtual, thus capturing the RNA conformational flexibility. Such an analysis has been made possible by RNAtango, a web server introduced in this paper, that provides a trigonometric perspective on RNA 3D structures, giving insights into the variability of examined models and their alignment with reference targets. RNAtango offers comprehensive tools for calculating torsion and pseudo-torsion angles, generating angle statistics, comparing RNA structures based on backbone torsions, and assessing local and global structural similarities using trigonometric functions and angle measures. The system operates in three scenarios: single model analysis, model-versus-target comparison, and model-versus-model comparison, with results output in text and graphical formats. Compatible with all modern web browsers, RNAtango is accessible freely along with the source code. It supports researchers in accurately assessing structural similarities, which contributes to the precision and efficiency of RNA modeling.

BCAN: Bidirectional Correct Attention Network for Cross-Modal Retrieval.

As a fundamental topic in bridging the gap between vision and language, cross-modal retrieval purposes to obtain the correspondences' relationship between fragments, i.e., subregions in images and words in texts. Compared with earlier methods that focus on learning the visual semantic embedding from images and sentences to the shared embedding space, the existing methods tend to learn the correspondences between words and regions via cross-modal attention. However, such attention-based approaches invariably result in semantic misalignment between subfragments for two reasons: 1) without modeling the relationship between subfragments and the semantics of the entire images or sentences, it will be hard for such approaches to distinguish images or sentences with multiple same semantic fragments and 2) such approaches focus attention evenly on all subfragments, including nonvisual words and a lot of redundant regions, which also will face the problem of semantic misalignment. To solve these problems, this article proposes a bidirectional correct attention network (BCAN), which introduces a novel concept of the relevance between subfragments and the semantics of the entire images or sentences and designs a novel correct attention mechanism by modeling the local and global similarity between images and sentences to correct the attention weights focused on the wrong fragments. Specifically, we introduce a concept about the semantic relationship between subfragments and entire images or sentences and use this concept to solve the semantic misalignment from two aspects. In our correct attention mechanism, we design two independent units to correct the weight of attention focused on the wrong fragments. Global correct unit (GCU) with modeling the global similarity between images and sentences into the attention mechanism to solve the semantic misalignment problem caused by focusing attention on relevant subfragments in irrelevant pairs (RI) and the local correct unit (LCU) consider the difference in the attention weights between fragments among two steps to solve the semantic misalignment problem caused by focusing attention on irrelevant subfragments in relevant pairs (IR). Extensive experiments on large-scale MS-COCO and Flickr30K show that our proposed method outperforms all the attention-based methods and is competitive to the state-of-the-art. Our code and pretrained model are publicly available at: https://github.com/liuyyy111/BCAN.

A Multiple Targets ISAR Imaging Method with Removal of Micro-Motion Connection Based on Joint Constraints

Combining multiple data sources, Digital Earth is an integrated observation platform based on air–space–ground–sea monitoring systems. Among these data sources, the Inverse Synthetic Aperture Radar (ISAR) is a crucial observation method. ISAR is typically utilized to monitor both military and civilian ships due to its all-day and all-weather superiority. However, in complex scenarios, multiple targets may exist within the same radar antenna beam, resulting in severe defocusing due to different motion conditions. Therefore, this paper proposes a multiple-target ISAR imaging method with the removal of micro-motion connections based on the integration of joint constraints. The fully motion-compensated targets exhibit low rank and local similarity in the high-resolution range profile (HRRP) domain, while the micro-motion components possess sparsity. Additionally, targets display sparsity in the image domain. Inspired by this, we formulate a novel optimization by promoting the low-rank, the Laplacian, and the sparsity constraints of targets and the sparsity constraints of the micro-motion components. This optimization problem is solved by the linearized alternative direction method with adaptive penalty (LADMAP). Furthermore, the different motions of various targets degrade their inherent characteristics. Therefore, we integrate motion compensation transformation into the optimization, accordingly achieving the separation of rigid bodies and the micro-motion components of different targets. Experiments based on simulated data demonstrate the effectiveness of the proposed method.

Double-Layer Distributed and Integrated Fault Detection Strategy for Non-Gaussian Dynamic Industrial Systems.

Currently, with the increasing scale of industrial systems, multisensor monitoring data exhibit large-scale dynamic Gaussian and non-Gaussian concurrent complex characteristics. However, the traditional principal component analysis method is based on Gaussian distribution and uncorrelated assumptions, which are greatly limited in practice. Therefore, developing a new fault detection method for large-scale Gaussian and non-Gaussian concurrent dynamic systems is one of the urgent challenges to be addressed. To this end, a double-layer distributed and integrated data-driven strategy based on Laplacian score weighting and integrated Bayesian inference is proposed. Specifically, in the first layer of the distributed strategy, we design a Jarque-Bera test module to divide all multisensor monitoring variables into Gaussian and non-Gaussian blocks, successfully solving the problem of different data distributions. In the second layer of the distributed strategy, we design a dynamic augmentation module to solve dynamic problems, a K-means clustering module to mine local similarity information of variables, and a Laplace scoring module to quantitatively evaluate the structural retention ability of variables. Therefore, this double-layer distributed strategy can simultaneously combine the different distribution characteristics, dynamism, local similarity, and importance of variables, comprehensively mining the local information of the multisensor data. In addition, we develop an integrated Bayesian inference strategy based on detection performance weighting, which can emphasize the differential contribution of local models. Finally, the fault detection results for the Tennessee Eastman production system and a diesel engine working system validate the superiority of the proposed method.

Experimental study on plunging breaking waves impacting a vertical cylinder with three different model scales

Scale effects, which can influence the reliability and effectiveness of experimental results, are often unavoidable but seldom explored, particularly in the context of plunging waves. To investigate the scale effects of plunging wave impacts on a vertical cylinder, experiments were conducted in a wave flume at model scales of 1:20, 1:30 and 1:40. The wavemaker movements were scaled based on the Froude-similarity law and the pressures on the front surface of the cylinder were analyzed. The results indicated that plunging waves were generated across these three scales, with fine agreement in overall wave elevations and a local waveform similarity of 80–90%. The analysis of pressures showed that the distribution of peak pressures on the cylinders remained consistent across different scales; however, scale effects were closely related to the intensity of wave impacts. Specifically, in cases of relatively weak impact, such as when the wavefront struck the cylinder before breaking or even after an intense wave breaking with intense gas-liquid mixtures, the pressure discrepancies were less than 20% among these different scales. This discrepancy can exceed 50% near the wave peaks during the vertical wave front or early jet formation stages with strong impacts.

Mixed convection flow and heat-mass transfer of cross liquid near a stagnation region with Dufour, Soret effects and chemical reaction

Heat-mass transfer in a liquid flow over a stretching surface at a stagnation regime is significant in optimal manufacturing and quality assurance depending on the rheological behavior of viscoelastic fluids in process machinery. Stretching sheet flow at a stagnation point is a standard procedure used in the metallurgical industries for manufacturing of different equipment and cooling systems. Due to its various applications in industrial operations, the current study aims to explore the heat-mass transfer of buoyancy-driven flow of cross liquid over a stretching sheet at a stagnation point. The heat and mass transfer under the framework of Dufour and Soret effects, radiant heat, and chemical reactions are assessed. Furthermore, the analysis also takes into account the linear and nonlinear convection-diffusion phenomena. The model findings are obtained by merging the Bvp4c in MATLAB using a local similarity technique up to the fourth truncation threshold. The findings illustrate that the action of magnetic field can be used to control the flow across the sheet. Temperature increases with thermal radiation, Biot number and Dufour number. The action of buoyancy ratio parameter result in improvement in fluid flow. With increasing trend of magnetic and Weissenberg number, the flow field and skin friction coefficient drops.

UT-1 Transporter Expression in the Spiny Dogfish (Squalus acanthias): UT-1 Protein Shows a Different Localization in Comparison to That of Other Sharks.

The original UT-1 transporter gene was initially identified in the spiny dogfish (Squalus acanthias), but localization of the UT-1 protein was not determined. Subsequent UT-1 expression was shown to localize to the collecting tubule (CT) of the shark nephron in other shark species, with expression in a closely related chimaera species also located additionally at a lower level in the intermediate-I segment (IS-I) of the nephron. In spiny dogfish, two UT-1 splice variants are known (UT-1 long and short), and there was also a second UT-1 gene described (here termed Brain UT). In this study, a second splice variant of the second Brain UT gene was discovered. Expression profiles (mRNA) of UT-1 long and short and Brain UT were determined in a number of spiny dogfish tissues. Quantitative PCR in kidney samples showed that the level of the short variant of UT-1 was around 100 times higher than the long variant, which was itself expressed around 10 times higher than Brain UT cDNA/mRNA (in kidney). For the long variant, there was a significantly higher level of mRNA abundance in fish acclimatized to 75% seawater. Ultimately, three UT-1 antibodies were made that could bind to both the UT-1 short and long variant proteins. The first two of these showed bands of appropriate sizes on Western blots of around 52.5 and 46 kDa. The second antibody had some additional lower molecular weight bands. The third antibody was mainly bound to the 46 kDa band with faint 52.5 kDa staining. Both the 52.5 and 46 kDa bands were absent when the antibodies were pre-blocked with the peptide antigens used to make them. Across the three antibodies, there were many similarities in localization but differences in subcellular localization. Predominantly, antibody staining was greatest in the intermediate segment 1 (IS-I) and proximal (PIb) segments of the first sinus zone loop of the nephron, with reasonably strong expression also found at the start and middle of the late distal tubule (LDT; second sinus zone loop). While some expression in the collecting tubule (CT) could not be ruled out, the level of staining seemed to be low or non-existent in convoluted bundle zone nephron segments such as the CT. Hence, this suggests that spiny dogfish have a fundamentally different mode of urea absorption in comparison to that found in other shark species, potentially focused more on the nephron sinus zone loops than the CT.

QIGTD: identifying critical genes in the evolution of lung adenocarcinoma with tensor decomposition

BackgroundIdentifying critical genes is important for understanding the pathogenesis of complex diseases. Traditional studies typically comparing the change of biomecules between normal and disease samples or detecting important vertices from a single static biomolecular network, which often overlook the dynamic changes that occur between different disease stages. However, investigating temporal changes in biomolecular networks and identifying critical genes is critical for understanding the occurrence and development of diseases.MethodsA novel method called Quantifying Importance of Genes with Tensor Decomposition (QIGTD) was proposed in this study. It first constructs a time series network by integrating both the intra and inter temporal network information, which preserving connections between networks at adjacent stages according to the local similarities. A tensor is employed to describe the connections of this time series network, and a 3-order tensor decomposition method was proposed to capture both the topological information of each network snapshot and the time series characteristics of the whole network. QIGTD is also a learning-free and efficient method that can be applied to datasets with a small number of samples.ResultsThe effectiveness of QIGTD was evaluated using lung adenocarcinoma (LUAD) datasets and three state-of-the-art methods: T-degree, T-closeness, and T-betweenness were employed as benchmark methods. Numerical experimental results demonstrate that QIGTD outperforms these methods in terms of the indices of both precision and mAP. Notably, out of the top 50 genes, 29 have been verified to be highly related to LUAD according to the DisGeNET Database, and 36 are significantly enriched in LUAD related Gene Ontology (GO) terms, including nuclear division, mitotic nuclear division, chromosome segregation, organelle fission, and mitotic sister chromatid segregation.ConclusionIn conclusion, QIGTD effectively captures the temporal changes in gene networks and identifies critical genes. It provides a valuable tool for studying temporal dynamics in biological networks and can aid in understanding the underlying mechanisms of diseases such as LUAD.

Unsupervised selective labeling for semi-supervised industrial defect detection

In industrial detection scenarios, achieving high accuracy typically relies on extensive labeled datasets, which are costly and time-consuming. This has motivated a shift towards semi-supervised learning (SSL), which leverages labeled and unlabeled data to improve learning efficiency and reduce annotation costs. This work proposes the unsupervised spectral clustering labeling (USCL) method to optimize SSL for industrial challenges like defect variability, rarity, and complex distributions. Integral to USCL, we employ the multi-task fusion self-supervised learning (MTSL) method to extract robust feature representations through multiple self-supervised tasks. Additionally, we introduce the Enhanced Spectral Clustering (ESC) method and a dynamic selecting function (DSF). ESC effectively integrates both local and global similarity matrices, improving clustering accuracy. The DSF maximally selects the most valuable instances for labeling, significantly enhancing the representativeness and diversity of the labeled data. USCL consistently improves various SSL methods compared to traditional instance selection methods. For example, it boosts Efficient Teacher by 5%, 6.6%, and 7.8% in mean Average Precision(mAP) on the Automotive Sealing Rings Defect Dataset, the Metallic Surface Defect Dataset, and the Printed Circuit Boards (PCB) Defect Dataset with 10% labeled data. Our work sets a new benchmark for SSL in industrial settings.

A method for image–text matching based on semantic filtering and adaptive adjustment

As image–text matching (a critical task in the field of computer vision) links cross-modal data, it has captured extensive attention. Most of the existing methods intended for matching images and texts explore the local similarity levels between images and sentences to align images with texts. Even though this fine-grained approach has remarkable gains, how to further mine the deep semantics between data pairs and focus on the essential semantics in data remains to be quested. In this work, a new semantic filtering and adaptive approach (FAAR) was proposed to ease the above problem. To be specific, the filtered attention (FA) module selectively focuses on typical alignments with the interference of meaningless comparisons eliminated. Next, the adaptive regulator (AR) further adjusts the attention weights of key segments for filtered regions and words. The superiority of our proposed method was validated by a number of qualitative experiments and analyses on the Flickr30K and MSCOCO data sets.

A Variant of the Local Similarity Theory and Approximations of Vertical Profiles of Turbulent Moments of the Atmospheric Convective Boundary Layer

The approximation of the turbulent moments of the atmospheric convective layer is based on a variant of the local similarity theory using the concepts of the semi-empirical theory of Prandtl turbulence. In the proposed variant of the local similarity theory, the second moment of vertical velocity and the “spectral” Prandtl mixing length are used as basic parameters. This approach allows us to extend Prandtl’s theory to turbulent moments of vertical velocity and buoyancy and additionally offer more than ten new approximations. The comparison of the proposed approximation with other variants of the theory of local similarity is considered. It is shown that the selected basic parameters significantly improve the agreement between the local similarity approximations and experimental data. The approximations are consistent with observations in the turbulent convective layer of the atmosphere, the upper boundary of which nearly corresponds to the lower boundary of the temperature inversion. Analytical approximations of local similarity can find applications in the construction of high-order moment closures in the vortex of resolving numerical turbulence models, as well as in the construction of “mass-flux” parametrization.

Partial label learning via weighted centroid clustering disambiguation

Partial Label Learning (PLL) is a weakly supervised learning problem that induces a multi-class classifier from data with candidate labels, among which only one is the ground-truth label. The crucial challenge in PLL is to disambiguate the false-positive labels from the candidate labels. However, most existing PLL methods fail to simultaneously consider both the instance-level similarity and the class-level information during label disambiguation. In this paper, we propose a novel two-stage method based on weighted centroid clustering, which efficiently utilizes both the local similarity among instances and the per-class global information. In the first stage, we initialize the center of each class using label propagation based on the instance-level similarity, and obtain the disambiguated instances via weighted centroid clustering derived from the per-class global information. In the second stage, the disambiguated instances are used to train a multi-class classifier. Extensive experiments on both controlled UCI datasets and real-world datasets show the superiority of the proposed method in classification accuracy.

An Edge-Preserving Regularization Model for the Demosaicing of Noisy Color Images

This paper proposes an edge-preserving regularization technique to solve the color image demosaicing problem in the realistic case of noisy data. We enforce intra-channel local smoothness of the intensity (low-frequency components) and inter-channel local similarity of the depth of object borders and textures (high-frequency components). Discontinuities of both the low-frequency and high-frequency components are accounted for implicitly, i.e., through suitable functions of the proper derivatives. For the treatment of even the finest image details, derivatives of first, second, and third orders are considered. The solution to the demosaicing problem is defined as the minimizer of an energy function, accounting for all these constraints plus a data fidelity term. This non-convex energy is minimized via an iterative deterministic algorithm, applied to a family of approximating functions, each implicitly referring to geometrically consistent image edges. Our method is general because it does not refer to any specific color filter array. However, to allow quantitative comparisons with other published results, we tested it in the case of the Bayer CFA and on the Kodak 24-image dataset, the McMaster (IMAX) 18-image dataset, the Microsoft Demosaicing Canon 57-image dataset, and the Microsoft Demosaicing Panasonic 500-image dataset. The comparisons with some of the most recent demosaicing algorithms show the good performance of our method in both the noiseless and noisy cases.

ST-SCSR: identifying spatial domains in spatial transcriptomics data via structure correlation and self-representation.

Recent advances in spatial transcriptomics (ST) enable measurements of transcriptome within intact biological tissues by preserving spatial information, offering biologists unprecedented opportunities to comprehensively understand tissue micro-environment, where spatial domains are basic units of tissues. Although great efforts are devoted to this issue, they still have many shortcomings, such as ignoring local information and relations of spatial domains, requiring alternatives to solve these problems. Here, a novel algorithm for spatial domain identification in Spatial Transcriptomics data with Structure Correlation and Self-Representation (ST-SCSR), which integrates local information, global information, and similarity of spatial domains. Specifically, ST-SCSR utilzes matrix tri-factorization to simultaneously decompose expression profiles and spatial network of spots, where expressional and spatial features of spots are fused via the shared factor matrix that interpreted as similarity of spatial domains. Furthermore, ST-SCSR learns affinity graph of spots by manipulating expressional and spatial features, where local preservation and sparse constraints are employed, thereby enhancing the quality of graph. The experimental results demonstrate that ST-SCSR not only outperforms state-of-the-art algorithms in terms of accuracy, but also identifies many potential interesting patterns.

A local space transfer learning-based parallel Bayesian optimization with its application

The optimization of process parameters in polyolefin production can bring significant economic benefits to the factory. However, due to small data sets, high costs associated with parameter verification cycles, and difficulty in establishing an optimization model, the optimization process is often restricted. To address this issue, we propose using a transfer learning Bayesian optimization strategy to improve the efficiency of parameter optimization while minimizing resource consumption. Specifically, we leverage Gaussian process (GP) regression models to establish an integrated model that incorporates both source and target grade production task data. We then measure the similarity weights of each model by comparing their predicted trends, and utilize these weights to accelerate the solution of optimal process parameters for producing target polyolefin grades. In order to enhance the accuracy of our approach, we acknowledge that measuring similarity in a global search space may not effectively capture local similarity characteristics. Therefore, we propose a novel method for transfer learning optimization that operates within a local space (LSTL-PBO). This method employs partial data acquired through random sampling from the target task data and utilizes Bayesian optimization techniques for model establishment. By focusing on a local search space, we aim to better discern and leverage the inherent similarities between source tasks and the target task. Additionally, we incorporate a parallel concept into our method to address multiple local search spaces simultaneously. By doing so, we can explore different regions of the parameter space in parallel, thereby increasing the chances of finding optimal process parameters. This localized approach allows us to improve the precision and effectiveness of our optimization process. The performance of our method is validated through experiments on benchmark problems, and we discuss the sensitivity of its hyperparameters. The results show that our proposed method can significantly improve the efficiency of process parameter optimization, reduce the dependence on source tasks, and enhance the method's robustness. This has great potential for optimizing processes in industrial environments.

SDA-FC: Bridging federated clustering and deep generative model

Federated clustering (FC) is an extension of centralized clustering in federated settings. The key here is how to construct a global similarity measure without sharing private data, since the local similarity may be insufficient to group local data correctly, and the similarity of samples across clients cannot be directly measured due to privacy constraints. Obviously, the most straightforward way to analyze FC is to employ methods extended from centralized ones, such as K-means (KM) and fuzzy c-means (FCM). However, they are vulnerable to non independent-and-identically-distributed (non-IID) data among clients. To handle this, we propose a pretty simple and effective federated clustering framework instantiated with generative adversarial network (GAN), named synthetic data aided federated clustering (SDA-FC). It trains generative adversarial network locally in each client and uploads the generated synthetic data to the server, where KM or FCM is performed on the synthetic data. The synthetic data can make the model immune to the non-IID problem and enable us to capture the global similarity characteristics more effectively without sharing private data. Comprehensive experiments reveal the advantages of SDA-FC, including superior performance in addressing the non-IID problem and the device failures. The code is available at https://github.com/Jarvisyan/SDA-FC.

SimMix: Local similarity-aware data augmentation for time series

We find that local similarity is an essential factor for data augmentation in deep learning tasks concerning time series data, the applications of which are prevalent in various domains such as smart healthcare, intelligent transportation, smart finance, etc. With empirical and theoretical analysis, we find deep learning models achieve excellent performance only when the data augmentation method performs with appropriate intensity of local similarity—during the data augmentation process, too large/small intra-class local similarity will decrease the performance of deep learning models. With this discovery, we propose a time series augmentation method based on intra-class Similarity Mixing (SimMix), which accurately controls the intensity by quantifying and adjusting the similarity between augmented samples and original samples. With a PAC (i.e., Probably Approximately Correct) theoretical foundation, we design a cutmix strategy for non-equal length segments to eliminate semantic information loss and noise introduction defects in traditional methods. Through extensive validation on 10 real-world datasets, we demonstrate that the proposed method can outperform the state-of-the-art by a large margin.

Phaeocystis globosa and diatom blooms promote distinct bacterial communities and associations in a coastal ecosystem.

Phytoplankton and bacteria form the foundation of marine food webs. While most studies on phytoplankton bloom influence on bacteria dynamics focus on diatom-dominated blooms due to their global ecological significance, it is unclear if similar patterns extend to other species that compete with diatoms like Phaeocystis spp. This study aimed to contribute to the understanding of associations between phytoplankton and bacteria in a temperate ecosystem. For this, we studied the dynamics of phytoplankton and bacteria, combining 16S metabarcoding, microscopy, and flow cytometry over 4 years (282 samples). Phytoplankton and bacterial communities were studied throughout the year, particularly during contrasting phytoplankton blooms dominated by the Haptophyte Phaeocystis globosa or diatoms. We applied extended local similarity analysis (eLSA) to construct networks during blooming and non-blooming periods. Overall, the importance of seasonal and species-specific interactions between phytoplankton and bacteria is highlighted. In winter, mixed diatom communities were interconnected with bacteria, indicating a synergistic degradation of diverse phytoplankton-derived substrates. In spring, despite the intensity variations of P. globosa blooms, the composition of bacterial communities remained consistent over several years, suggesting establishing a stable-state environment for bacterial communities. Specific associations between monospecific diatom blooms and bacteria were evidenced in summer.