Energy-efficient UAV communication: A NOMA scheme with resource allocation and trajectory optimization.

This work investigates a downlink nonorthogonal multiple access (NOMA) scheme with unmanned aerial vehicle (UAV) aided wireless communication, where a single UAV was regarded as an air base station (ABS) to communicate with multiple ground users. Considering the constraints of velocity and maneuverability, a UAV energy efficiency (EE) model was proposed via collaborative design resource allocation and trajectory optimization. Based on this, an EE maximization problem was formulated to jointly optimize the transmit power of ground users and the trajectory of the UAV. To obtain the optimal solutions, this nonconvex problem was transformed into an equivalent convex optimization problem on the basis of three user clustering algorithms. After several alternating iterations, our proposed algorithms converged quickly. The simulation results show an enhancement in EE with NOMA because our proposed algorithm is nearly 99.6% superior to other OMA schemes.

PWLU: Learning Specialized Activation Functions With the Piecewise Linear Unit.

The choice of activation functions is crucial to deep neural networks. ReLU is a popular hand-designed activation function. Swish, the automatically searched activation function, outperforms ReLU on many challenging datasets. However, the search method has two main drawbacks. First, the tree-based search space is highly discrete and restricted, which is difficult to search. Second, the sample-based search method is inefficient in finding specialized activation functions for each dataset or neural architecture. To overcome these drawbacks, we propose a new activation function called Piecewise Linear Unit (PWLU), incorporating a carefully designed formulation and learning method. PWLU can learn specialized activation functions for different models, layers, or channels. Besides, we propose a non-uniform version of PWLU, which maintains sufficient flexibility but requires fewer intervals and parameters. Additionally, we generalize PWLU to three-dimensional space to define a piecewise linear surface named 2D-PWLU, which can be treated as a non-linear binary operator. Experimental results show that PWLU achieves SOTA performance on various tasks and models, and 2D-PWLU is better than element-wise addition when aggregating features from different branches. The proposed PWLU and its variation are easy to implement and efficient for inference, which can be widely applied in real-world applications.

Theoretical Studies of Mutual Effects between 6-m-r Hemiketalization and 26-m-r Lactonization in Pimaricin Thioesterase.

Pimaricin is a small polyene macrolide antibiotic and has been broadly used as an antimycotic and antiprotozoal agent in both humans and foods. As a thioesterase in type-I polyketide synthase, pimTE controls the 26-m-r macrolide main chain release in pimaricin biosynthesis. In this work, we sought to determine whether the 6-m-r hemiketal formation was linked to pimTE-catalyzed 26-m-r lactonization. Compared to non-hemiketal TEs, pimTE is characterized by an aspartic acid residue (D179) accessible to the U-turn motif in the acyl-enzyme intermediate. Both the covalent docking and molecular dynamics simulations demonstrate that the reactive conformations for macrocyclic lactonization are drastically promoted by the 6-m-r hemiketal. Moreover, the small-model quantum mechanistic calculations suggest that protic residues can significantly accelerate the 6-m-r hemiketal cyclization. In addition, the post-hemiketal molecular dynamic simulations demonstrate that hydrogen-bonding networks surrounding the substrate U-turn of the hairpin-shaped conformation changes significantly when the 6-m-r hemiketal is formed. In particular, the R-hemiketal intermediate is not only catalyzed by the D179 residue, but also twists the hairpin structure to the 26-m-r lactonizing pre-reaction state. By contrast, the S-hemiketal formation is unlikely catalyzed by D179, which twists the hairpin in an opposite direction. Our results propose that pimTE could be a bi-functional enzyme, which can synergistically catalyze tandem 6-m-r and 26-m-r formations during the main-chain release of pimaricin biosynthesis.

Research progress on deep learning algorithms to assist 3D tooth segmentation of digital dental models / 口腔疾病防治

@#Three-dimensional tooth segmentation is the segmentation of single-tooth models from a digital dental model. It is an important foundation for diagnosis, planning, treatment and customized appliance manufacturing in digital orthodontics. With the deep integration of artificial intelligence technology and big data from stomatology, the use of deep learning algorithms to assist 3D tooth segmentation has gradually become mainstream. This review summarizes the current situation of deep learning algorithms that assist 3D tooth segmentation from the aspects of dataset establishment, algorithm architecture, algorithm performance, innovation and advantages, deficiencies of current research and prospects. The results of the literature review showed that deep learning tooth segmentation methods could obtain an accuracy of more than 95% and had good robustness. However, the segmentation of complex dental models, operation time and richness of the training database still need to be improved. Research and development of the "consumption reduction and strong core" algorithm, establishment of an authoritative data sample base with multiple centers, and expansion of data application depth and breadth will lead to further development in this field.

Endowing homodimeric carbamoyltransferase GdmN with iterative functions through structural characterization and mechanistic studies.

Iterative enzymes, which catalyze sequential reactions, have the potential to improve the atom economy and diversity of industrial enzymatic processes. Redesigning one-step enzymes to be iterative biocatalysts could further enhance these processes. Carbamoyltransferases (CTases) catalyze carbamoylation, an important modification for the bioactivity of many secondary metabolites with pharmaceutical applications. To generate an iterative CTase, we determine the X-ray structure of GdmN, a one-step CTase involved in ansamycin biosynthesis. GdmN forms a face-to-face homodimer through unusual C-terminal domains, a previously unknown functional form for CTases. Structural determination of GdmN complexed with multiple intermediates elucidates the carbamoylation process and identifies key binding residues within a spacious substrate-binding pocket. Further structural and computational analyses enable multi-site enzyme engineering, resulting in an iterative CTase with the capacity for successive 7-O and 3-O carbamoylations. Our findings reveal a subclade of the CTase family and exemplify the potential of protein engineering for generating iterative enzymes.

Involvement of IL-10R/STAT3 pathway in amyloid ß clearance by microlgia in Alzheimer's disease.

Both the total amount and annual growth rate of Alzheimer's disease (AD) patients in China are much higher than in other regions in the world. This trend of rapid growth will be difficult to change in the next few decades, hence the prevention and treatment situation of AD patients in China is more severe. Maintaining the balance between the production and removal pathways of Aß is an important guarantee for the body to maintain its normal physiological state. The dysfunction of Aß clearance is an important factor of Aß accumulation in brain tissue of AD patients causing neurotoxicity of synaptic damage and neuronal death. Based on the literature review, it introduced the important role of microglias in clearing Aß deposits in the process of Alzheimer's disease. And most of these phagocytic cells were the specific phenotype of disease-related microglia (DAM-I/DAM-II) that induced microglial differentiation after activation. IL-10KO promoted the transformation of microglial phenotype DAM-II, and enhanced its phagocytosis for Aß oligomers. There is a hypothesis that IL-10R/STAT3 negatively regulates microglial phagocytosis. It was learnt that blocking the IL-10R/STAT3 pathway promoted microglial activation and enhanced phagocytosis. The comprehensive review on the involvement of IL-10R/STAT3 pathway in the process of AD would open up new ideas and discover new targets for the development of new therapeutic drugs.

The Role of Neutrophil Extracellular Traps in Periodontitis.

Periodontitis is a chronic, destructive disease of periodontal tissues caused by multifaceted, dynamic interactions. Periodontal bacteria and host immunity jointly contribute to the pathological processes of the disease. The dysbiotic microbial communities elicit an excessive immune response, mainly by polymorphonuclear neutrophils (PMNs). As one of the main mechanisms of PMN immune response in the oral cavity, neutrophil extracellular traps (NETs) play a crucial role in the initiation and progression of late-onset periodontitis. NETs are generated and released by neutrophils stimulated by various irritants, such as pathogens, host-derived mediators, and drugs. Chromatin and proteins are the main components of NETs. Depending on the characteristics of the processes, three main pathways of NET formation have been described. NETs can trap and kill pathogens by increased expression of antibacterial components and identifying and trapping bacteria to restrict their spread. Moreover, NETs can promote and reduce inflammation, inflicting injuries on the tissues during the pro-inflammation process. During their long-term encounter with NETs, periodontal bacteria have developed various mechanisms, including breaking down DNA of NETs, degrading antibacterial proteins, and impacting NET levels in the pocket environment to resist the antibacterial function of NETs. In addition, periodontal pathogens can secrete pro-inflammatory factors to perpetuate the inflammatory environment and a friendly growth environment, which are responsible for the progressive tissue damage. By learning the strategies of pathogens, regulating the periodontal concentration of NETs becomes possible. Some practical ways to treat late-onset periodontitis are reducing the concentration of NETs, administering anti-inflammatory therapy, and prescribing broad-spectrum and specific antibacterial agents. This review mainly focuses on the mechanism of NETs, pathogenesis of periodontitis, and potential therapeutic approaches based on interactions between NETs and periodontal pathogens.

Mouse strain specificity of DAAO inhibitors-mediated antinociception.

D-Amino acid oxidase (DAAO) specifically catalyzes the oxidative deamination of neutral and polar D-amino acids and finally yields byproducts of hydrogen peroxide. Our previous work demonstrated that the spinal astroglial DAAO/hydrogen peroxide (H2 O2 ) pathway was involved in the process of pain and morphine antinociceptive tolerance. This study aimed to report mouse strain specificity of DAAO inhibitors on antinociception and explore its possible mechanism. DAAO inhibitors benzoic acid, CBIO, and SUN significantly inhibited formalin-induced tonic pain in Balb/c and Swiss mice, but had no antinociceptive effect in C57 mice. In contrast, morphine and gabapentin inhibited formalin-induced tonic pain by the same degrees among Swiss, Balb/c and C57 mice. Therefore, mouse strain difference in antinociceptive effects was DAAO inhibitors specific. In addition, intrathecal injection of D-serine greatly increased spinal H2 O2 levels by 80.0% and 56.9% in Swiss and Balb/c mice respectively, but reduced spinal H2 O2 levels by 29.0% in C57 mice. However, there was no remarkable difference in spinal DAAO activities among Swiss, Balb/c and C57 mice. The spinal expression of glutathione (GSH) and glutathione peroxidase (GPx) activity in C57 mice were significantly higher than Swiss and Balb/c mice. Furthermore, the specific GPx inhibitor D-penicillamine distinctly restored SUN antinociception in C57 mice. Our results reported that DAAO inhibitors produced antinociception in a strain-dependent manner in mice and the strain specificity might be associated with the difference in spinal GSH and GPx activity.

Immunomodulation of Chinese Herbal Medicines on NK cell populations for cancer therapy: A systematic review.

ETHNOPHARMACOLOGICAL RELEVANCE: Immunomodulation has become a crucial modality for cancer treatment. Chinese Herbal Medicines (CHMs) are expected as adjuvant therapy for immunomodulation against cancer, but face the key challenge of poor scientific evidence. Changes of natural killer (NK) cells on numbers and/or cytotoxicity are a novel respect to evaluate the immunomodulation of CHMs. AIM OF THE STUDY: The purpose of this review is to investigate the immunomodulation of Chinese Herbal Medicines (CHMs) on NK cell populations for cancer therapy. MATERIALS AND METHODS: A systematic review was conducted and outside mainstream electronic databases were screened for potential reference articles. This review tried to report and critically analyzed all the correlative studies, especially these clinical trials (3 CHM extracts and 11 CHM formulas). RESULTS: Evidence-based functions of CHMs against cancer could be summarized as: (1) enhancement of NK cells activity or relative percentage; (2) prevention of tumor growth and metastasis; (3) relief on side-effects or complications of therapeutic strategies (i.e. chemotherapy, radiotherapy and resection). Briefly, most of cellular studies and two thirds animal studies were based on the extract or components of single herbs, whilst most of clinical trials were keen on formula or prescription of CHMs. The main components of CHMs were demonstrated active on promoting the cytotoxicity of NK cells, including Angelica sinensis, Ganoderma lucidum, Panax ginseng, Radix Astragali, Lentinus edodes, etc. CONCLUSIONS: This comprehensive review demonstrated NK cells activity was positively associated with quality of life but not survival benefit of cancer patients. Thus exploring the roles of NK cells in adjuvant therapy against cancer is confirmed to be beneficial to explore the underlying relationship between immunomodulation and quality of life.

Crafting GBD-Net for Object Detection.

The visual cues from multiple support regions of different sizes and resolutions are complementary in classifying a candidate box in object detection. Effective integration of local and contextual visual cues from these regions has become a fundamental problem in object detection. In this paper, we propose a gated bi-directional CNN (GBD-Net) to pass messages among features from different support regions during both feature learning and feature extraction. Such message passing can be implemented through convolution between neighboring support regions in two directions and can be conducted in various layers. Therefore, local and contextual visual patterns can validate the existence of each other by learning their nonlinear relationships and their close interactions are modeled in a more complex way. It is also shown that message passing is not always helpful but dependent on individual samples. Gated functions are therefore needed to control message transmission, whose on-or-offs are controlled by extra visual evidence from the input sample. The effectiveness of GBD-Net is shown through experiments on three object detection datasets, ImageNet, Pascal VOC2007 and Microsoft COCO. Besides the GBD-Net, this paper also shows the details of our approach in winning the ImageNet object detection challenge of 2016, with source code provided on https://github.com/craftGBD/craftGBD. In this winning system, the modified GBD-Net, new pretraining scheme and better region proposal designs are provided. We also show the effectiveness of different network structures and existing techniques for object detection, such as multi-scale testing, left-right flip, bounding box voting, NMS, and context.

Performance Optimization of Marine Science and Numerical Modeling on HPC Cluster.

Marine science and numerical modeling (MASNUM) is widely used in forecasting ocean wave movement, through simulating the variation tendency of the ocean wave. Although efforts have been devoted to improve the performance of MASNUM from various aspects by existing work, there is still large space unexplored for further performance improvement. In this paper, we aim at improving the performance of propagation solver and data access during the simulation, in addition to the efficiency of output I/O and load balance. Our optimizations include several effective techniques such as the algorithm redesign, load distribution optimization, parallel I/O and data access optimization. The experimental results demonstrate that our approach achieves higher performance compared to the state-of-the-art work, about 3.5x speedup without degrading the prediction accuracy. In addition, the parameter sensitivity analysis shows our optimizations are effective under various topography resolutions and output frequencies.

Directed accumulation of less toxic pimaricin derivatives by improving the efficiency of a polyketide synthase dehydratase domain.

Pimaricin is an important polyene antifungal antibiotic that binds ergosterol and extracts it from fungal membranes. In previous work, two pimaricin derivatives (1 and 2) with improved pharmacological activities and another derivative (3) that showed no antifungal activity were produced by the mutant strain of Streptomyces chattanoogensis, in which the P450 monooxygenase gene scnG has been inactivated. Furthermore, inactivation of the DH12 dehydratase domain of the pimaricin polyketide synthases (PKSs) resulted in specific accumulation of the undesired metabolite 3, suggesting that improvement of the corresponding dehydratase activity may reduce or eliminate the accumulation of 3. Accordingly, the DH12-KR12 didomain within the pimaricin PKS was swapped with the fully active DH11-KR11 didomain. As predicted, the mutant was not able to produce 3 but accumulated 1 and 2 in high yields. Moreover, the effect of the flanking linker regions on domain swapping was evaluated. It was found that retention of the DH12-KR12 linker regions was more critical for the processivity of hybrid PKSs. Subsequently, high-yield production of 1 or 2 was obtained by overexpressing the scnD gene and its partner scnF and by disrupting the scnD gene, respectively. To our knowledge, this is the first report on the elimination of a polyketide undesired metabolite along with overproduction of desired product by improving the catalytic efficiency of a DH domain using a domain swapping technology.

Coherent Terahertz Radiation from Multiple Electron Beams Excitation within a Plasmonic Crystal-like structure.

Coherent terahertz radiation from multiple electron beams excitation within a plasmonic crystal-like structure (a three-dimensional holes array) which is composed of multiple stacked layers with 3 × 3 subwavelength holes array has been proposed in this paper. It has been found that in the structure the electromagnetic fields in each hole can be coupled with one another to construct a composite mode with strong field intensity. Therefore, the multiple electron beams injection can excite and efficiently interact with such mode. Meanwhile, the coupling among the electron beams is taken place during the interaction so that a very strong coherent terahertz radiation with high electron conversion efficiency can be generated. Furthermore, due to the coupling, the starting current density of this mechanism is much lower than that of traditional electron beam-driven terahertz sources. This multi-beam radiation system may provide a favorable way to combine photonics structure with electronics excitation to generate middle, high power terahertz radiation.

Gbps terahertz external modulator based on a composite metamaterial with a double-channel heterostructure.

The past few decades have witnessed a substantial increase in terahertz (THz) research. Utilizing THz waves to transmit communication and imaging data has created a high demand for phase and amplitude modulation. However, current active THz devices, including modulators and switches, still cannot meet THz system demands. Double-channel heterostructures, an alternative semiconductor system, can support nanoscale two-dimensional electron gases (2DEGs) with high carrier concentration and mobility and provide a new way to develop active THz devices. In this Letter, we present a composite metamaterial structure that combines an equivalent collective dipolar array with a double-channel heterostructure to obtain an effective, ultrafast, and all-electronic grid-controlled THz modulator. Electrical control allows for resonant mode conversion between two different dipolar resonances in the active device, which significantly improves the modulation speed and depth. This THz modulator is the first to achieve a 1 GHz modulation speed and 85% modulation depth during real-time dynamic tests. Moreover, a 1.19 rad phase shift was realized. A wireless free-space-modulation THz communication system based on this external THz modulator was tested using 0.2 Gbps eye patterns. Therefore, this active composite metamaterial modulator provides a basis for the development of effective and ultrafast dynamic devices for THz wireless communication and imaging systems.