Search

Acer palmatum (A. palmatum), a deciduous shrub or small arbour which belongs to Acer of Aceraceae, is an excellent greening species as well as a beautiful ornamental plant. In this study, a high-quality chromosome-level reference genome for A. palmatum was constructed using Oxford Nanopore sequencing and Hi-C technology. The assembly genome was ∼745.78 Mb long with a contig N50 length of 3.20 Mb, and 95.30 % (710.71 Mb) of the assembly was anchored into 13 pseudochromosomes. A total of 28,559 protein-coding genes were obtained, ∼90.02 % (25,710) of which could be functionally annotated. The genomic evolutionary analysis revealed that A. palmatum is most closely related to A. yangbiense and A. truncatum, and underwent only an ancient gamma whole-genome duplication event. Despite lacking a recent independent WGD, 25,795 (90.32 %) genes of A. palmatum were duplicated, and the unique/expanded gene families were linked with genes involved in plant-pathogen interaction and several metabolic pathways, which might underpin adaptability. A combined genomic, transcriptomic, and metabolomic analysis related to the biosynthesis of anthocyanin in leaves during the different season were characterized. The results indicate that the dark-purple colouration of the leaves in spring was caused by a high amount of anthocyanins, especially delphinidin and its derivatives; and the red colouration of the leaves in autumn by a high amount of cyanidin 3-O-glucoside. In conclusion, these valuable multi-omic resources offer important foundations to explore the molecular regulation mechanism in leaf colouration and also provide a platform for the scientific and efficient utilization of A. palmatum.

The rising mental health difficulties of the urban population in developing countries may be attributed to the high levels of air pollution. However, nationwide large-scale empirical works that examine this claim are rare. In this study, we construct a daily mental health metric using the volume of mental-health-related queries on the largest search engine in China, Baidu, to test this hypothesis. We find that air pollution causally undermines people’s mental health and that this impact becomes stronger as the duration of exposure to air pollution increases. Heterogeneity analyses reveal that men, middle-aged people and married people are more vulnerable to the impact of air pollution on mental health. More importantly, the results also demonstrate that the cumulative effects of air pollution on mental health are smaller for people living in cities with a higher gross domestic product per capita, more health resources, larger areas of green land and more sports facilities. Finally, we estimate that with a one-standard-deviation increase of fine particulate matter (26.3 μg m−3), the number of people who suffer from mental health problems in China increases by approximately 1.15 million. Our findings provide quantitative evidence for the benefits of reducing air pollution to promote mental health and well-being.

Competition between Points of Interest (POIs) refers to the situation in which two POIs directly or indirectly provide similar services to secure businesses. A large portion of prior studies on competition analysis focuses on mining textual data, e.g., news articles and social comments. However, the increasing availability of human mobility and mobile query data enables a new paradigm for analyzing the competitive relationships among POIs, which remains largely unexplored. To this end, in this paper, we attempt to mine large-scale online map search query data for better understanding POI competitive relationships. Based on a co-query POI graph built from the map search query data, we develop a novel neural graphlet-based prediction framework to predict the competitive relationships among POIs. A unique perspective of our model is to infer latent POI competitive relationships by integrating multiple distinct factors, e.g., graphlet structure, geographical distance, and regional features, reflected in map search query data and POI data. Finally, we conduct extensive experiments on real-world datasets to demonstrate the effectiveness of the proposed framework, and show that our framework outperforms all baselines with a significant margin in all evaluation metrics.

The ability to predict city-wide parking availability is crucial for the successful development of Parking Guidance and Information (PGI) systems. Indeed, the effective prediction of city-wide parking availability can improve parking efficiency, help urban planning and ultimately alleviate city congestion. However, it is a non-trivial task because of three major challenges:1) the non-Euclidean spatial autocorrelation among parking lots, 2) the dynamic temporal autocorrelation inside of and between parking lots, and 3) the scarcity of real-time parking availability information obtained from real-time sensors. To this end, we propose Semi-supervised Hierarchical Recurrent Graph Neural Network-X(SHARE-X) for predicting city-wide parking availability. Specifically, we first propose a hierarchical graph convolution structure to model non-Euclidean spatial autocorrelation among parking lots. Along this line, a contextual graph convolution block and a multi-resolution soft clustering graph convolution block are respectively proposed to capture local and global spatial dependencies between parking lots. Moreover, we devise a hierarchical attentive recurrent network module to incorporate both short and long-term dynamic temporal dependencies of parking lots. Additionally, a parking availability approximation module is introduced to estimate missing real-time parking availabilities from both spatial and temporal domains. Finally,experiments on two real-world datasets demonstrate the prediction performance of SHARE-X outperforms eight state-of-the-art baselines.

Modeling the global dynamics of emerging infectious diseases (EIDs) like COVID-19 can provide important guidance in the preparation and mitigation of pandemic threats. While age-structured transmission models are widely used to simulate the evolution of EIDs, most of these studies focus on the analysis of specific countries and fail to characterize the spatial spread of EIDs across the world. Here, we developed a global pandemic simulator that integrates age-structured disease transmission models across 3,157 cities and explored its usage under several scenarios. We found that without mitigations, EIDs like COVID-19 are highly likely to cause profound global impacts. For pandemics seeded in most cities, the impacts are equally severe by the end of the first year. The result highlights the urgent need for strengthening global infectious disease monitoring capacity to provide early warnings of future outbreaks. Additionally, we found that the global mitigation efforts could be easily hampered if developed countries or countries near the seed origin take no control. The result indicates that successful pandemic mitigations require collective efforts across countries. The role of developed countries is vitally important as their passive responses may significantly impact other countries.

Protein representation learning methods have shown great potential to yield useful representation for many downstream tasks, especially on protein classification. Moreover, a few recent studies have shown great promise in addressing insufficient labels of proteins with self-supervised learning methods. However, existing protein language models are usually pretrained on protein sequences without considering the important protein structural information. To this end, we propose a novel structure-aware protein self-supervised learning method to effectively capture structural information of proteins. In particular, a well-designed graph neural network (GNN) model is pretrained to preserve the protein structural information with self-supervised tasks from a pairwise residue distance perspective and a dihedral angle perspective, respectively. Furthermore, we propose to leverage the available protein language model pretrained on protein sequences to enhance the self-supervised learning. Specifically, we identify the relation between the sequential information in the protein language model and the structural information in the specially designed GNN model via a novel pseudo bi-level optimization scheme. Experiments on several supervised downstream tasks verify the effectiveness of our proposed method.The code of the proposed method is available in \url{https://github.com/GGchen1997/STEPS_Bioinformatics}., Comment: Accepted by Bioinformatics; 7 pages 4 figures

User identification across multiple datasets has a wide range of applications and there has been an increasing set of research works on this topic during recent years. However, most of existing works focus on user identification with a single input data type, e.g., (I) identifying a user across multiple social networks with online data and (II) detecting a single user from heterogeneous trajectory datasets with offline data. Different from previous works, in this paper, we propose a framework on user identification between online and offline datasets. We build connections between these two types of data by a mapping from IP addresses to physical locations. To solve this problem, we propose a novel framework consists of three steps. First, we use a clustering method based on locations of IP addresses to map IP addresses into specific physical location distributions. Second, we propose a novel pairwise index to reduce space cost and running time for computing the co-occurrence. Lastly, we apply a learning-to-rank method to merge the effect of multiple features we get in the first two steps. Based on our framework, we design experiments to demonstrate the efficiency (in time and space) of our framework, together with the precision and recall of our approach compared to other methods.

Effective molecular representation learning is of great importance to facilitate molecular property prediction. Recent advances for molecular representation learning have shown great promise in applying graph neural networks to model molecules. Moreover, a few recent studies design self-supervised learning methods for molecular representation to address insufficient labelled molecules; however, these self-supervised frameworks treat the molecules as topological graphs without fully utilizing the molecular geometry information. The molecular geometry, also known as the three-dimensional spatial structure of a molecule, is critical for determining molecular properties. To this end, we propose a novel geometry-enhanced molecular representation learning method (GEM). The proposed GEM has a specially designed geometry-based graph neural network architecture as well as several dedicated geometry-level self-supervised learning strategies to learn the molecular geometry knowledge. We compare GEM with various state-of-the-art baselines on different benchmarks and show that it can considerably outperform them all, demonstrating the superiority of the proposed method.

The transmission of certain bacteria from the mother's gut to the infant's gut via breast milk (BM) is critical for offspring's immune system development. Dysbiosis of the BM microbiota can be caused by a variety of reasons, which can be influenced by probiotics delivered via the enteromammary route. The goal of this study was to investigate the bacterium that can be transmitted from mother to infant's intestine during various lactation periods in 19 mother-child dyads. Bacterial transmission is most common during the colostrum phase when bacteria with certain amplicon sequence variants (ASVs) enter the newborn intestine and inhabit it permanently. We have established that anaerobic gut-associated bacteria, such as Faecalibacterium, Blautia and Lachnoclostridium, transfer from the mother to the infant's gut with lactation dependence using the idea of weighted transfer ratios. Streptococcus salivarius, Bifidobacterium longum, and Lactobacillus gasseri were transferred from the maternal gut to the BM, as well as from the BM to the newborn gut, dependent on different ASVs. These findings suggest that isolating key microorganisms from breast milk could be utilized to modify the microbiota of BM or newborns by giving the mother a probiotic or adding it to artificial milk to promote neonatal health.

The consumption of mixed nuts is a healthy dietary strategy to reduce the risk of cardiovascular disease and has a prebiotic effect on the gut microbiota. However, there is a lack of basic research based on mixed nut formulation. This study established a new method for optimizing mixed nut formulations using the Nutrient Rich Food (NRF) index model. Nutrient indices were adjusted by combining 10 and 8 encouraging nutrients and 3 limiting nutrients of nuts and dried fruits, respectively. The optimized mixed nut formulation had the highest total NRF and the lowest energy, which was achieved by applying linear programming. The effect of an optimized mixed nut formulation on insulin resistance and gut microbiota was investigated in an animal model of metabolic disorders caused by a high-fat diet. Male C57BL/6J mice (

Gut microbiota transmission from mother to offspring is critical to infant gut microbiota and immune development. Mother's intestine and breast milk are rich in secretory immunoglobulin A (sIgA), which can coat a specific bacterial spectrum and may be related to bacterial transmission and colonization. Here we analyzed the microbiota and sIgA-coated bacteria of maternal fecal samples and breast milk and infant fecal samples from 19 dyads by 16S rRNA amplicon sequencing. For the sIgA-coated microbiota, both the phylogenetic diversity and the Shannon index of maternal fecal samples show a lower trend than those of infant fecal samples (

Different nanostructured CoP microcubes derived from metal formate frameworks with enhanced oxygen evolution reaction performance.

Graph Neural Networks (GNNs) have become widely-used models for semi-supervised learning. However, the robustness of GNNs in the presence of label noise remains a largely under-explored problem. In this paper, we consider an important yet challenging scenario where labels on nodes of graphs are not only noisy but also scarce. In this scenario, the performance of GNNs is prone to degrade due to label noise propagation and insufficient learning. To address these issues, we propose a novel RTGNN (Robust Training of Graph Neural Networks via Noise Governance) framework that achieves better robustness by learning to explicitly govern label noise. More specifically, we introduce self-reinforcement and consistency regularization as supplemental supervision. The self-reinforcement supervision is inspired by the memorization effects of deep neural networks and aims to correct noisy labels. Further, the consistency regularization prevents GNNs from overfitting to noisy labels via mimicry loss in both the inter-view and intra-view perspectives. To leverage such supervisions, we divide labels into clean and noisy types, rectify inaccurate labels, and further generate pseudo-labels on unlabeled nodes. Supervision for nodes with different types of labels is then chosen adaptively. This enables sufficient learning from clean labels while limiting the impact of noisy ones. We conduct extensive experiments to evaluate the effectiveness of our RTGNN framework, and the results validate its consistent superior performance over state-of-the-art methods with two types of label noises and various noise rates., 9 pages, accepted to WSDM 2023 Research Track

Left unmitigated, climate change poses a catastrophic risk to human health, demanding an urgent and concerted response from every country. The 2015 Lancet Commission on Health and Climate Change and The Lancet Countdown: Tracking Progress on Health and Climate Change have been initiated to map out the impacts of climate change and the necessary policy responses. To meet these challenges, Tsinghua University, partnering with the University College London and 17 Chinese and international institutions, has prepared the Chinese Lancet Countdown report, which has a national focus and builds on the work of the global Lancet Countdown: Tracking Progress on Health and Climate Change. Drawing on international methodologies and frameworks, this report aims to deepen the understanding of the links between public health and climate change at the national level and track them with 23 indicators. This work is part of the Lancet's Countdown broader efforts to develop regional expertise on this topic, and coincides with the launch of the Lancet Countdown Regional Centre in Asia, based at Tsinghua University. The data and results of this report are presented at the provincial level, where possible, to facilitate targeted response strategies for local decision-makers. Based on the data and findings of the 2020 Chinese Lancet Countdown report, five recommendations are proposed to key stakeholders in health and climate change in China: (1) Enhance inter-departmental cooperation. Climate change is a challenge that demands an integrated response from all sectors, urgently requiring substantial inter-departmental cooperation among health, environment, energy, economic, financial, and education authorities. (2) Strengthen health emergency preparedness. Knowledge and findings on current and future climate-related health threats still lack the required attention and should be fully integrated into the emergency preparedness and response system. (3) Support research and raise awareness. Additional financial support should be allocated to health and climate change research in China to enhance health system adaptation, mitigation measures, and their health benefits. At the same time, media and academia should be fully motivated to raise the public and politicians' awareness of this topic. (4) Increase climate change mitigation. Speeding up the phasing out of coal is necessary to be consistent with China's pledge to be carbon neutral by 2060 and to continue to reduce air pollution. Fossil fuel subsidies must also be phased out. (5) Ensure the recovery from COVID-19 to protect health now and in the future. China's efforts to recover from COVID-19 will shape public health for years to come. Climate change should be a priority in these interventions.

Diabetic nephropathy is a serious complication of diabetes. Hyperoside has been widely reported to ameliorate diabetes-associated disease. The current study is designed to explore the mechanism of hyperoside in diabetic nephropathy. In the present study, high glucose was used to treat podocytes. Diabetic nephropathy mice models were established by high-fat feeding followed by multiple low dose injections of streptozocin. Western blot analysis was conducted for detection of extracellular matrix accumulation, inflammatory response and cell apoptosis. We found out that hyperoside improved high glucose-induced cell injury. Additionally, hyperoside prevented mice with diabetic nephropathy from diabetic symptoms and renal dysfunction. Mechanistically, hyperoside inhibited the mRNA and protein expression of APC. MiR-499-5p was found to be an upstream negative mediator of APC, and hyperoside induced the upregulation of miR-499-5p. MiR-499-5p bound with the 3' untranslated region of APC to inhibit its expression. Finally, rescue assays revealed that the suppressive effects of miR-499-5p overexpression on renal dysfunction were rescued by upregulation of APC in mice with diabetic nephropathy. In conclusion, these findings indicated that hyperoside ameliorates diabetic nephropathy via targeting the miR-499-5p/APC axis, suggesting that hyperoside may offer a potential tactic for diabetic nephropathy treatment.

Changes in secretory immunoglobulin A (SIgA) coated bacteria from early to late pregnancy were associated with the development of gestational diabetes mellitus (GDM). SIgA coated beneficial gut bacteria, which are depleted in GDM, are potential probiotics for the prevention of GDM. We investigated blood biochemistry, chronic inflammation, mucosal barrier biomarkers and faecal SIgA coated microbiota in healthy early pregnancy (T1H, n = 50), late pregnancy (T3H, n = 30) and women with GDM (T3D, n = 27). The “leaky gut” markers, zonulin and lipopolysaccharide (LPS), significantly increased in T3D compared to the T3H group. The Shannon index of SIgA coated microbiota was elevated in late pregnancy compared to early pregnancy and was the highest in the T3D group (p < 0.001). The T3D group was enriched in SIgA coated Escherichia and Streptococcus and depleted in Lactobacillus and Bifidobacterium. Blood glucose (BG) positively correlated with zonulin (p < 0.001) and LPS (p < 0.05). Lactobacillus reuteri negatively correlated with BG (p < 0.05), zonulin (p < 0.05) and LPS (p < 0.01). Lactobacillus reuteri QS01 isolated from the feces of T1H significantly reduced LPS released by the gut microbiota of GDM individuals in vitro. In conclusion, GDM may be related to intestinal mucosal damage and inflammation-induced dysbiosis of SIgA coated microbiota. SIgA coated L. reuteri can reduce the level of LPS of GDM in vitro.

Considering the rigid-flexible coupling system of 5 MW wind turbine, the static pressure, fluid temperature, Mach number, vorticity, shear stress and velocity distribution of tower flange are obtained under the combined action of wind shear and fluctuating wind. The dynamic response of rigid-flexible coupling structure is studied by the combined action of the whole vibration and local vibration of 5 MW wind turbine. Through the analysis of dynamic parameter change curve, the improvement measures of tower structure are put forward, and the dynamic response of the structure before and after improvement is compared. The superiority of the improved structure is expounded, so as to provide reference basis for engineering application.

Multi-modal transportation recommendation aims to provide the most appropriate travel route with various transportation modes according to certain criteria. After analyzing large-scale navigation data, we find that route representations exhibit two patterns: spatio-temporal autocorrelations within transportation networks and the semantic coherence of route sequences. However, there are few studies that consider both patterns when developing multi-modal transportation systems. To this end, in this paper, we study multi-modal transportation recommendation with unified route representation learning by exploiting both spatio-temporal dependencies in transportation networks and the semantic coherence of historical routes. Specifically, we propose to unify both dynamic graph representation learning and hierarchical multi-task learning for multi-modal transportation recommendations. Along this line, we first transform the multi-modal transportation network into time-dependent multi-view transportation graphs and propose a spatiotemporal graph neural network module to capture the spatial and temporal autocorrelation. Then, we introduce a coherent-aware attentive route representation learning module to project arbitrary-length routes into fixed-length representation vectors, with explicit modeling of route coherence from historical routes. Moreover, we develop a hierarchical multi-task learning module to differentiate route representations for different transport modes, and this is guided by the final recommendation feedback as well as multiple auxiliary tasks equipped in different network layers. Extensive experimental results on two large-scale real-world datasets demonstrate the performance of the proposed system outperforms eight baselines.

Human mobility annotation aims to assign mobility records the corresponding visiting Point-of-Interests (POIs). It is one of the most fundamental problems for understanding human mobile behaviors. In literature, many efforts have been devoted to annotating mobility records in a pointwise or trajectory-wise manner. However, the user preference factor is not fully explored and, worse still, the mobile peer influence factor has never been integrated. To this end, in this article, we propose a novel framework, named JEPPI, to jointly exploit user preference and mobile peer influence to tackle the problem. In our JEPPI, we first unify the two distinct factors in a behavior-driven user-POI graph. This graph enables us to model user preference with user-POI visiting relationships, and model two types of mobile peer influence with co-location and co-visiting peer relationships, respectively. Moreover, we devise an equivalence-emphasizing metric to reduce redundancy in the second-order co-visiting peer influence. In addition, a mutual augmentation learning approach is proposed to preserve the latent structures of various factors exploited. Notably, our learning approach preserves all factors in a shared representation space such that user preference is learned with mobile peer influence being considered at the same time, and vice versa. In this way, the different factors are mutually augmented and semantically integrated to enhance human mobility annotation. Finally, using two large-scale real-world datasets, we conduct extensive experiments to demonstrate the superiority of our approach compared with the state-of-the-art annotation methods.

The ability to predict city-wide parking availability is crucial for the successful development of Parking Guidance and Information (PGI) systems. Indeed, the effective prediction of city-wide parking availability can improve parking efficiency, help urban planning, and ultimately alleviate city congestion. However, it is a non-trivial task for predicting citywide parking availability because of three major challenges: 1) the non-Euclidean spatial autocorrelation among parking lots, 2) the dynamic temporal autocorrelation inside of and between parking lots, and 3) the scarcity of information about real-time parking availability obtained from real-time sensors (e.g., camera, ultrasonic sensor, and GPS). To this end, we propose Semi-supervised Hierarchical Recurrent Graph Neural Network (SHARE) for predicting city-wide parking availability. Specifically, we first propose a hierarchical graph convolution structure to model non-Euclidean spatial autocorrelation among parking lots. Along this line, a contextual graph convolution block and a soft clustering graph convolution block are respectively proposed to capture local and global spatial dependencies between parking lots. Additionally, we adopt a recurrent neural network to incorporate dynamic temporal dependencies of parking lots. Moreover, we propose a parking availability approximation module to estimate missing real-time parking availabilities from both spatial and temporal domain. Finally, experiments on two real-world datasets demonstrate the prediction performance of SHARE outperforms seven state-of-the-art baselines., Comment: 8 pages, 9 figures, AAAI-2020

Automatic diagnosis systems have attracted increasing attention in recent years. The reinforcement learning (RL) is an attractive technique for building an automatic diagnosis system due to its advantages for handling sequential decision making problem. However, the RL method still cannot achieve good enough prediction accuracy. In this paper, we propose a Generative Adversarial regularized Mutual information Policy gradient framework (GAMP) for automatic diagnosis which aims to make a diagnosis rapidly and accurately. We first propose a new policy gradient framework based on the Generative Adversarial Network (GAN) to optimize the RL model for automatic diagnosis. In our framework, we take the generator of GAN as a policy network, and also use the discriminator of GAN as a part of the reward function. This generative adversarial regularized policy gradient framework can try to avoid generating randomized trials of symptom inquires deviated from the common diagnosis paradigm. In addition, we add mutual information to enhance the reward function to encourage the model to select the most discriminative symptoms to make a diagnosis. Experiment evaluations on two public datasets show that our method beats the state-of-art methods, not only can achieve higher diagnosis accuracy, but also can use a smaller number of inquires to make diagnosis decision.

Objectives:Lactobacillus reuteri Fn041 (Fn041) is a probiotic isolated from immunoglobulin A coated microbiota in the human breast milk of Gannan in China with a low incidence of hypercholesterolemia. This study aims to explore the role and mechanism of Fn041 in preventing hypercholesterolemia caused by a high-fat diet in mice.MethodsC57BL/6N mice were fed a low-fat diet or a high-fat diet and gavage with Fn041 and Lactobacillus rhamnosus GG (LGG) for 8 weeks.ResultsBoth Fn041 and LGG prevented the occurrence of hypercholesterolemia, liver and testicular fat accumulation. In addition, a high-fat diet causes intestinal dysbiosis and mucosal barrier damage, which is associated with hypercholesterolemia. Fn041 prevented the high-fat diet-induced reduction in alpha diversity of intestinal microbiota and intestinal mucosal barrier damage. Fn041 treatment significantly increased fecal total cholesterol and total bile acids.ConclusionsFn041 prevented hypercholesterolemia by enhancing cholesterol excretion and mucosal barrier function.

C57BL/6N mice were fed a low-fat diet or a high-fat diet and gavage with Fn041 andBoth Fn041 and LGG prevented the occurrence of hypercholesterolemia, liver and testicular fat accumulation. In addition, a high-fat diet causes intestinal dysbiosis and mucosal barrier damage, which is associated with hypercholesterolemia. Fn041 prevented the high-fat diet-induced reduction in alpha diversity of intestinal microbiota and intestinal mucosal barrier damage. Fn041 treatment significantly increased fecal total cholesterol and total bile acids.Fn041 prevented hypercholesterolemia by enhancing cholesterol excretion and mucosal barrier function.

The Lauraceae is a family of the order Laurales, with 2500–3000 species comprising 50 genera, mainly distributed in tropical and subtropical evergreen broad-leaved forests. Up to two decades ago, the systematic classification of the Lauraceae was based on floral morphology, but molecular phylogenetic approaches have made tremendous advances in elucidating tribe- and genus-level relationships within the family in recent decades. Our review focused on the phylogeny and systematics of Sassafras, a genus of three species with highly disjunct distributions in eastern North America and East Asia, whose tribe affiliation within the Lauraceae has long been controversial. By synthesizing information on the floral biology and molecular phylogeny of Sassafras, this review aimed to explore the position of Sassafras within the Lauraceae, and to provide suggestions and implications for future phylogenetic studies. Our synthesis highlighted Sassafras as a transitional type between Cinnamomeae and Laureae with a closer genetic relationship with Cinnamomeae, as revealed by molecular phylogenetic evidence, while it shares many similar characteristics with Laureae in morphology. We therefore discovered that several molecular and morphological methods should be concurrently considered to illuminate the phylogeny and systematics of Sassafras in Lauraceae.

While online advertising is one of the major sources of income for search engines, pumping up the incomes from business advertisements while ensuring the user experience becomes a challenging but emerging area. Designing high-quality advertisements with persuasive content has been proved as a way to increase revenues through improving the Click-Through Rate (CTR). However, it is difficult to scale up the design of high-quality ads, due to the lack of automation in creativity. In this paper, we present Commonsense-Enriched Advertisement on Search Engine (CHASE) --- a system for the automatic generation of persuasive ads. CHASE adopts a specially designed language model that fuses the keywords, commonsense-related texts, and marketing contents to generate persuasive advertisements. Specifically, the language model has been pre-trained using massive contents of explicit knowledge and fine-tuned with well-constructed quasi-parallel corpora with effective control of the proportion of commonsense in the generated ads and fitness to the ads' keywords. The effectiveness of the proposed method CHASE has been verified by real-world web traffics for search and manual evaluation. In A/B tests, the advertisements generated by CHASE would bring 11.13% CTR improvement. The proposed model has been deployed to cover three advertisement domains (which are kid education, psychological counseling, and beauty e-commerce) at Baidu, the world's largest Chinese search engine, with adding revenue of about 1 million RMB (Chinese Yuan) per day.

Supplementing Limosilactobacillus reuteri Fn041, a breast milk-derived probiotic from agricultural and pastoral areas, to maternal mice during late pregnancy and lactation prevents atopic dermatitis (AD) in offspring. This study aims to elucidate the molecular mechanism of Fn041-mediated immune regulation.Fn041 is administered prenatal and postnatal to maternal mice, and to offspring after weaning. The ears are administered with calcipotriol to induce AD. Fn041 treatment significantly alleviates ear inflammation, and reduces mast cell infiltration. Fn041 treatment upregulates and downregulates intestinal ZO-1 and Claudin-2 mRNA expression, respectively. Transcriptome analysis of Peyer's patches reveals that pathways related to DNA damage repair are activated in AD mice, which is inhibited by Fn041 treatment. Fn041 activates pathways related to retinol absorption and metabolism. Untargeted metabolomic analysis reveals that Fn041 treatment increases plasma retinol and kynurenine. Fn041 treatment does not significantly alter the overall cecal microbiota profile, only increases the relative abundances of Ligilactobacillus apodemi, Ligilactobacillus murinus, Akkermansia muciniphila, and Bacteroides thetaiotaomicron.Fn041 induces anti-AD immune responses directly by promoting the absorption and metabolism of retinol in Peyer's patches, and plays an indirect role by strengthening the mucosal barrier and increasing the abundance of specific anti-AD bacteria in the cecum.

Peer reviewed: True, Rural living environment governance is an issue of great importance in global economic development. It is also a significant facet of the rural revitalization strategy pursued by the Chinese government. Henan Province is a major agricultural province in the inland Central Plains region and is one of the most populous provinces in China. In this study, we take two villages in the underdeveloped areas of Henan Province in central China as case studies. We use field investigations and in-depth interviews to analyze the feelings of farmers in the case study villages towards the efforts to improve the rural environment, and explored the factors that affect levels of satisfaction with the rural living environment from the perspective of grassroots governance. The results showed that there were significant differences in the perceptions and satisfaction levels of rural residents in the two villages. The grassroots government has great influence over the rural human settlement environment, and its deployment capacities and management level are the key factors that affect overall levels of farmer satisfaction. China’s experience in improving rural human settlements provides reference for rural governance in other developing countries and regions around the world.

Sassafras tzumu (Hemsl.) Hemsl., as an Arctic Tertiary relict woody species, is an ecologically and economically important deciduous tree endemic to southern China. Nonetheless, the genetic resources and backgrounds of S. tzumu are still lacking and remain largely unclear. Here, we predicted 16,215 candidate polymorphic nuclear microsatellite (nSSR) loci from the assembled nucleus databases of six geographic-distant individuals of S. tzumu via CandiSSR. Among these nSSRs, the di- (75.53%) and tri-nucleotide (19.75%) repeats were the most abundant, and 27 new polymorphic SSRs were developed and characterized in 136 individuals from six natural populations of S. tzumu. The majority of the above 27 SSRs (24 loci, 88.89%) presented moderate polymorphism (mean PIC = 0.356), and the transferability of these markers in other Sassafras species was high (85.19%). A moderately low level of genetic diversity and a high variation (FST = 0.286) of six wild populations of S. tzumu were illuminated by 16 selected polymorphic nSSRs, with the average expected heterozygosity (HE) of 0.430 at the species level and HE ranging from 0.195 to 0.387 at the population level. Meanwhile, a bottleneck effect was shown in two populations. Consistent with the results of the principal coordinate analysis (PCoA) and phylogenetic trees, structure analysis optimally divided these six S. tzumu populations into two clusters, and the further strong population subdivision appeared from K = 2 to K = 5, which corresponded to two evolutionarily significant units (ESUs). Moreover, the significant correlation between genetic and geographic distance was tested by the Mantel test (r = 0.742, p = 0.006), clarifying the effect about isolation by distance (IBD), which could be possibly explained by the low gene flow (Nm = 0.625), a relatively high degree of inbreeding (FIS = 0.166), a relatively large distribution, and mountainous barriers. Above all, our research not only enlarged the useful genetic resources for future studies of population genetics, molecular breeding, and germplasm management of S. tzumu and its siblings but also contributed to proposing scientific conservation strategies and schemes for the better preservation of S. tzumu and other Sassafras (Lauraceae) species.

China implemented the first phase of its National Healthy Cities pilot program from 2016-20. Along with related urban health governmental initiatives, the program has helped put health on the agenda of local governments while raising public awareness. Healthy City actions taken at the municipal scale also prepared cities to deal with the COVID-19 pandemic. However, after intermittent trials spanning the past two decades, the Healthy Cities initiative in China has reached a crucial juncture. It risks becoming inconsequential given its overlap with other health promotion efforts, changing public health priorities in response to the pandemic, and the partial adoption of the Healthy Cities approach advanced by the World Health Organization (WHO). We recommend aligning the Healthy Cities initiative in China with strategic national and global level agendas such as Healthy China 2030 and the Sustainable Development Goals (SDGs) by providing an integrative governance framework to facilitate a coherent intersectoral program to systemically improve population health. Achieving this alignment will require leveraging the full spectrum of best practices in Healthy Cities actions and expanding assessment efforts.Tsinghua-Toyota Joint Research Fund "Healthy city systems for smart cities" program.

Effective molecular representation learning is of great importance to facilitate molecular property prediction, which is a fundamental task for the drug and material industry. Recent advances in graph neural networks (GNNs) have shown great promise in applying GNNs for molecular representation learning. Moreover, a few recent studies have also demonstrated successful applications of self-supervised learning methods to pre-train the GNNs to overcome the problem of insufficient labeled molecules. However, existing GNNs and pre-training strategies usually treat molecules as topological graph data without fully utilizing the molecular geometry information. Whereas, the three-dimensional (3D) spatial structure of a molecule, a.k.a molecular geometry, is one of the most critical factors for determining molecular physical, chemical, and biological properties. To this end, we propose a novel Geometry Enhanced Molecular representation learning method (GEM) for Chemical Representation Learning (ChemRL). At first, we design a geometry-based GNN architecture that simultaneously models atoms, bonds, and bond angles in a molecule. To be specific, we devised double graphs for a molecule: The first one encodes the atom-bond relations; The second one encodes bond-angle relations. Moreover, on top of the devised GNN architecture, we propose several novel geometry-level self-supervised learning strategies to learn spatial knowledge by utilizing the local and global molecular 3D structures. We compare ChemRL-GEM with various state-of-the-art (SOTA) baselines on different molecular benchmarks and exhibit that ChemRL-GEM can significantly outperform all baselines in both regression and classification tasks. For example, the experimental results show an overall improvement of 8.8% on average compared to SOTA baselines on the regression tasks, demonstrating the superiority of the proposed method., Comment: Nature Machine Intelligence, 2022

Medical entity relation verification is a crucial step to build a practical and enterprise medical knowledge graph (MKG) because high-precision medical entity relation is a key requirement for many MKG-based applications. Existing relation verification approaches for general knowledge graphs are not designed for considering medical domain knowledge, although it is central to achieve high-quality entity relation verification for MKG. To this end, in this paper, we introduce a system for medical entity relation verification with large-scale machine reading comprehension. The proposed system is tailored to overcome the unique challenges of medical relation verification including high variants of medical terms, the high difficulty of evidence searching in complex medical documents, and the lack of evidence labels for supervision. To deal with the problem of variants of medical terms, we introduce a synonym-aware retrieve model to retrieve the potential evidence implicitly verifying the given claim. To better utilize the medical domain knowledge, a relation-aware evidence detector and a medical ontology-enhanced aggregator are developed to improve the performance of the relation verification module. Moreover, to overcome the challenge of providing high-quality evidence due to the lack of labels, we introduce an interactive collaborative-training method to iteratively improve the evidence accuracy. Finally, we conduct extensive experiments to demonstrate that the performance of our proposed system is superior to all comparable models. We also demonstrate that our system can significantly reduce the annotation time by medical experts in real-world verification tasks. It can help to improve the efficiency by nearly 300%. In particular, our system has been embedded into the Baidu Clinical Decision Support System.

Logic locking is necessary for protecting intellectual property. Although logic-locking schemes have been proposed to resist the powerful satisfiability (SAT)-based attack, many of them are subject to removal attacks, which identify the logic- locking block and replace its output by the correct signal, so the circuit would function correctly without the right key. In order to prevent removal attacks, the stripped functional logic locking-Hamming distance (SFLL-HD) scheme and its variations corrupt the original circuit and add a logic-locking block to correct the errors. The high-complexity HD checker can be also replaced by the Cascaded(CAS)-lock block. This paper proposes a new low-complexity logic-locking scheme inspired by the Generalized(G)-Anti-SAT block. By relaxing the constraint of the G-Anti-SAT design in the SFLL setting, our new logic-locking scheme reduces the logic complexity by around 37% compared to the M-CAS block with similar resistance to the SAT and removal attacks. Additionally, unlike the SFLL-HD and Mirrored (M)- CAS schemes, the proposed logic-locking block can use a large variation of functions. This makes existing or potential attacks utilizing properties of the logic function impossible.

In order to measure and reconstruct accurate three-dimension (3D) data for visual aided navigation of autonomous land vehicles (ALVs), a multimedia stereo calibration algorithm which is suitable for normal scene and especially for low illumination scene is proposed. Firstly, an expression of object-point re-projection errors is derived by the collinear equation model, and the non-linear least square algorithm (NLS) is introduced to iteratively optimize external parameters for individual camera. A rectangular pyramidal method enforcing the rectangular geometric constraint is presented, to produce more stable initial parameter values. Then, according to imaging-point correspondences between the left and right camera, a re-projection error model is constructed for this stereo calibration system, of which all parameters are further optimized and calculated through the calibrated results of two separate cameras. Experimental results show that the proposed algorithm can achieve re-projection errors of no more than 0.5 pixels and converge fast usually with less than 10 interation times, whether under normal illumination or low illumination, so it can get better performance and realize a rapid re-calibration.

The toxic effects of lead on human health and the environment have long been a focus of research. To explore sources of lead in Guangzhou, China, we investigated atmospheric lead-containing particles (LCPs) during wintertime using a single particle aerosol mass spectrometer (SPAMS). Based on mass spectral features, LCPs were classified into eight major particle types, including Pb-Cl and Pb-Cl-Li (coal combustion and waste incineration), Pb-Cl-EC and Pb-Cl-OC (diesel trucks and coal combustion), Pb-Cl-Fe (iron and steel industry), Pb-Cl-AlSi (dust), Pb-Sec (secondary formation), and Pb-Cl-Zn (industrial process); these sources (in parentheses) were identified by comparing atmospheric LCP mass spectra with authentic Pb emission source mass spectra. Sampling periods with LCP number fractions (NFs) more than three times the average LCP NF (APF = 4.35%) and below the APF were defined as high LCP NF periods (HLFPs: H1, H3, and H5) and low LCP NF APF periods (LLFPs: L2 and L4), respectively. Diurnal patterns and high Pb-Sec content during LLFPs indicate that photochemical activity and heterogeneous reactions may have controlled Pb-Sec particle formation. The inverse Pb-Cl and Pb-Sec particle diurnal trends during LLFPs suggest the replacement of Cl by sulfate and nitrate. On average over the five periods, ~ 76% of the LCPs likely arose from coal combustion and/or waste incineration, which were dominant sources during all five periods, followed by diesel trucks during LLFPs and iron- and steel-related sources during HLFPs; HLFP LCPs arose mainly from primary emissions. These results can be used to more efficiently control Pb emission sources and prevent harm to human and environmental health from Pb toxicity.

Hyperoside, a flavonol glycoside, is derived from plants of the genera Hypericum and Crataegus. Recent studies have indicated the anti-apoptotic and anti-inflammatory roles of hyperoside. The present study was designed to measure the effects of hyperoside on high glucose (HG)-treated HK-2 cells. HK-2 is a human papillomavirus 16 transformed cell line and can be used as a model for normal tubular cell. Cell apoptosis was examined by TUNEL assays and flow cytometry analysis. Inflammatory response was detected by Enzyme linked immunosorbent assay kits. Western blotting was applied to detect protein levels of apoptosis-related genes and inflammatory cytokines. Mechanistical assays including luciferase reporter and RNA pull down assays were applied to detect the binding relationship between molecules. We identified that hyperoside protected HK-2 cells against HG-induced apoptosis and inflammation. Moreover, miR-499a-5p was upregulated by hyperoside in a dose dependent manner. MiR-499a-5p inhibition rescued the suppressive effects of hyperoside on apoptosis and inflammation of HG-treated HK-2 cells. Furthermore, miR-499a-5p targeted NRIP1 to inhibit its mRNA expression, and further suppressed its translation. NRIP1 was downregulated by hyperoside in a dose dependent manner. Finally, rescue assays indicated that miR-499a-5p inhibition rescued the protective effects of hyperoside on apoptosis and inflammatory response of HK-2 cells by NRIP1. In conclusion, our findings revealed that hyperoside alleviates HG-induced apoptosis and inflammatory response of HK-2 cells by the miR-499a-5p/NRIP1 axis.