Recognition of cyanobacteria promoters via Siamese network-based contrastive learning under novel non-promoter generation

It is a vital step to recognize cyanobacteria promoters on a genome-wide scale. Computational methods are promising to assist in difficult biological identification. When building recognition models, these methods rely on non-promoter generation to cope with the lack of real non-promoters. Nevertheless, the factitious significant difference between promoters and non-promoters causes over-optimistic prediction. Moreover, designed for E. coli or B. subtilis, existing methods cannot uncover novel, distinct motifs among cyanobacterial promoters. To address these issues, this work first proposes a novel non-promoter generation strategy called phantom sampling, which can eliminate the factitious difference between promoters and generated non-promoters. Furthermore, it elaborates a novel promoter prediction model based on the Siamese network (SiamProm), which can amplify the hidden difference between promoters and non-promoters through a joint characterization of global associations, upstream and downstream contexts, and neighboring associations w.r.t. k-mer tokens. The comparison with state-of-the-art methods demonstrates the superiority of our phantom sampling and SiamProm. Both comprehensive ablation studies and feature space illustrations also validate the effectiveness of the Siamese network and its components. More importantly, SiamProm, upon our phantom sampling, finds a novel cyanobacterial promoter motif ('GCGATCGC'), which is palindrome-patterned, content-conserved, but position-shifted.

Apoptotic bodies inhibit inflammation by PDL1-PD1-mediated macrophage metabolic reprogramming

Apoptosis triggers immunoregulation to prevent and suppress inflammation and autoimmunity. However, the mechanism by which apoptotic cells modulate immune responses remains largely elusive. In the context of allogeneic mesenchymal stem cells (MSCs) transplantation, long-term immunoregulation is observed in the host despite the short survive of the injected MSCs. In this study, utilizing a mouse model of acute lung injury (ALI), we demonstrate that apoptotic bodies (ABs) released by transplanted human umbilical cord MSCs (UC-MSCs) convert the macrophages from a pro-inflammatory to an anti-inflammatory state, thereby ameliorating the disease. Mechanistically, we identify the expression of programmed cell death 1 ligand 1 (PDL1) on the membrane of UC-MSCs-derived ABs, which interacts with programmed cell death protein 1 (PD1) on host macrophages. This interaction leads to the reprogramming of macrophage metabolism, shifting from glycolysis to mitochondrial oxidative phosphorylation via the Erk-dependent pathway in ALI. Importantly, we have reproduced the PDL1-PD1 effects of ABs on metabolic switch using alveolar macrophages from patients with ALI, suggesting the potential clinical implications of developing therapeutic strategies for the patients.

Spatial-Temporal Self-Attention Enhanced Graph Convolutional Networks for Fitness Yoga Action Recognition

Fitness yoga is now a popular form of national fitness and sportive physical therapy. At present, Microsoft Kinect, a depth sensor, and other applications are widely used to monitor and guide yoga performance, but they are inconvenient to use and still a little expensive. To solve these problems, we propose spatial-temporal self-attention enhanced graph convolutional networks (STSAE-GCNs) that can analyze RGB yoga video data captured by cameras or smartphones. In the STSAE-GCN, we build a spatial-temporal self-attention module (STSAM), which can effectively enhance the spatial-temporal expression ability of the model and improve the performance of the proposed model. The STSAM has the characteristics of plug-and-play so that it can be applied in other skeleton-based action recognition methods and improve their performance. To prove the effectiveness of the proposed model in recognizing fitness yoga actions, we collected 960 fitness yoga action video clips in 10 action classes and built the dataset Yoga10. The recognition accuracy of the model on Yoga10 achieves 93.83%, outperforming the state-of-the-art methods, which proves that this model can better recognize fitness yoga actions and help students learn fitness yoga independently.

[Three-dimensional analysis of maxillary dentition during molar distalization with clear aligners under different movement designs: an in vitro experiment]

Objective: To investigate the three-dimensional force in the maxillary dentition under different movement designs for molar distalization with clear aligners Methods: Three groups were designed: simultaneous movement group (simultaneous distalization of maxillary first and second molars), second molar movement group (distalization of maxillary second molars) and first molar movement group (distalization of maxillary first molars). Ten clear aligners were made in each group, and the displacement was designed to be 0.2 mm. A force sensing device was established to measure the three-dimensional force on the upper dentition with the clear aligner. The device contained a model of the maxillary dentition consisting of 14 teeth, each tooth connected to an individual sensor. After the clear aligner was fitted, the data of 14 sensors were collected and analyzed using computer analysis software. The moving teeth were taken as the target teeth, and the rest of the teeth were anchorage. The data of the three-dimensional force in the three groups in each tooth position were measured and compared. Results: The sagittal forces on the first and second molars in the simultaneous movement group were (5.61±0.94) and (5.81±1.08) N, respectively, which were significantly smaller than those of the target teeth in the same position in other groups (P<0.05). The second molars in the first molar movement group received a sagittal reaction force, which was (-6.73±1.99) N. The anterior teeth in the three groups were all subjected to sagittal reaction force, and the force value was in a range of (-3.33 to 0.46) N. In the coronal direction, the second premolars of the simultaneous movement group received the reaction force in the palatal direction, and the force value was (-2.17±1.06) N. The first molars in the second molar movement group were also subjected to palatal reaction force of (-1.99±0.70) N. The second molars and second premolars in the first molar movement group were also subjected to palatal reaction force, which were (-2.85±0.57) and (-1.85±0.74) N, respectively. Compared with the sagittal and coronal forces, the target teeth and anchorage teeth in the three groups were less stressed in the vertical direction. Conclusions: The first and second molars distalized simultaneously, the correction force in the sagittal direction was relatively small. When first molar was moved distally alone, a greater reaction force in the sagittal direction was exerted on the second molar. Buccal displacement of the adjacent anchorage teeth should be designed to counteract the palatal reaction force on the anchorage teeth as the molars moved distally.