Search Results (4)

In video human action-recognition tasks, motion tempo describes the dynamic patterns and temporal scales of human motion. Different categories of actions are typically composed of sub-actions with varying motion tempos. Effectively capturing sub-actions with different motion tempos and distinguishing category-specific sub-actions are crucial for improving action-recognition performance. Convolutional Neural Network (CNN)-based methods attempted to address this challenge, by embedding feedforward attention modules to enhance the action’s dynamic representation learning. However, feedforward attention modules rely only on local information from low-level features, lacking contextual information to generate attention weights. Therefore, we propose a Sub-action Motion information Enhancement Network (SMEN) based on motion-tempo learning and feedback attention, which consists of the Multi-Granularity Adaptive Fusion Module (MgAFM) and Feedback Attention-Guided Module (FAGM). MgAFM enhances the model’s ability to capture crucial sub-action intrinsic information by extracting and adaptively fusing motion dynamic features at different granularities. FAGM leverages high-level features that contain contextual information in a feedback manner to guide low-level features in generating attention weights, enhancing the model’s ability to extract more discriminative spatio-temporal and channel-wise features. Experiments are conducted on three datasets, and the proposed SMEN achieves top-1 accuracies of 52.4%, 63.3% on the Something-Something V1 and V2 datasets, and 76.9% on the Kinetics-400 dataset. Ablation studies, evaluations, and visualizations demonstrate that the proposed SMEN is effective for sub-action motion tempo and representation learning, and outperforms compared methods for video action recognition. Full article

Virus-like particles (VLPs) are non-infectious and serve as promising vaccine platforms because they mimic the membrane-embedded conformations of fusion glycoproteins on native viruses. Here, we employed SARS-CoV-2 VLPs (SMEN) presenting ancestral, Beta, or Omicron spikes to identify the variant spike that elicits potent and cross-protective immune responses in the highly sensitive K18-hACE2 challenge mouse model. A combined intranasal and intramuscular SMEN vaccine regimen generated the most effective immune responses to significantly reduce disease burden. Protection was primarily mediated by antibodies, with minor but distinct contributions from T cells in reducing virus spread and inflammation. Immunization with SMEN carrying ancestral spike resulted in 100, 75, or 0% protection against ancestral, Delta, or Beta variant-induced mortality, respectively. However, SMEN with an Omicron spike provided only limited protection against ancestral (50%), Delta (0%), and Beta (25%) challenges. By contrast, SMEN with Beta spikes offered 100% protection against the variants used in this study. Thus, the Beta variant not only overcame the immunity produced by other variants, but the Beta spike also elicited diverse and effective humoral immune responses. Our findings suggest that leveraging the Beta variant spike protein can enhance SARS-CoV-2 immunity, potentially leading to a more comprehensive vaccine against emerging variants. Full article

The rapid evolution of new SARS-CoV-2 variants poses a continuing threat to human health. Vaccination has become the primary therapeutic intervention. The goal of the current work was the construction of immunogenic virus-like particles (VLPs). Here, we describe a human cell line for cost-efficient and scalable production of immunogenic SARS-CoV-2 VLPs. The modular design of the VLP-production platform facilitates rapid adaptation to new variants. Methods: The N, M-, and E-protein genes were integrated into the genome of Expi293 cells (ExpiVLP_MEN). Subsequently, this cell line was further modified for the constitutive expression of the SARS-CoV-2 spike protein. The resulting cell line (ExpiVLP_SMEN) released SARS-CoV-2 VLP upon exposure to doxycycline. ExpiVLP_SMEN cells were readily adapted for VLP production in a 5 L bioreactor. Purified VLPs were quantified by Western blot, ELISA, and nanoparticle tracking analysis and visualized by electron microscopy. Immunogenicity was tested in mice. Results: The generated VLPs contained all four structural proteins, are within the size range of authentic SARS-CoV-2 virus particles, and reacted strongly and specifically with immunoserum from naturally infected individuals. The VLPs were stable in suspension at 4 °C for at least 10 weeks. Mice immunized with VLPs developed neutralizing antibodies against lentiviruses pseudotyped with the SARS-CoV-2 spike protein. The flexibility of the VLP-production platform was demonstrated by the rapid switch of the spike protein to a new variant of concern (BA.1/Omicron). The present study describes an efficient, scalable, and adaptable production method of immunogenic SARS-CoV-2 VLPs with therapeutic potential. Full article

In Algeria, Smen is a fermented butter produced in households using empirical methods. Smen fermentation is driven by autochthonous microorganisms; it improves butter shelf-life and yields highly fragrant products used as ingredients in traditional dishes as well as in traditional medicine. The present study is aimed at investigating microbial diversity and dynamics during Algerian Smen fermentation using both culture-dependent and culture-independent approaches, as well as by monitoring volatile organic compound production. To reach this goal, fifteen Smen samples (final products) produced in households from different regions in Algeria were collected and analyzed. In addition, microbial and volatile compound dynamics at the different stages of Smen manufacturing were investigated for one Smen preparation. The results showed that Smen is a microbiologically safe product, as all hygiene and safety criteria were respected. The dominant microorganisms identified by both techniques were LAB and yeasts. Lactococcus spp. and Streptococcus thermophilus were the main bacterial species involved in spontaneous raw milk fermentation preceding butter-making, while lactobacilli and enterococci were the only bacteria found to be viable during Smen maturation. Regarding fungal diversity, yeast species were only recovered from two mature Smen samples by culturing, while different species (e.g., Geotrichum candidum, Moniliella sp.) were identified in all samples by the culture-independent approach. Using microbial analysis of a single batch, many of these were found viable during manufacturing. Concerning the volatile profiles, they were highly diverse and characterized by a high prevalence of short chain fatty acids, methylketones, and esters. Correlation analysis between microbial diversity and volatile profiles showed that several yeast (Moniliella sp., K. marxianus) and LAB (e.g., Lactococcus spp., S. thermophilus) species were strongly correlated with one or more volatile organic compound families, including several ethyl esters and methyl ketones that can be linked to pleasant, sweetly floral, fruity, buttery, and creamy odors. This study clearly identified key microorganisms involved in Smen fermentation and maturation that could be used in the future for better fermentation control and improvement of quality attributes. Full article