Search

Point-of-interest (POI) recommendation which aims at predicting the locations that users may be interested in has attracted wide attentions due to the development of Internet of Things and location-based services. Although collaborative filtering based methods and deep neural network have gain great success in POI recommendation, data sparsity and cold start problem still exist. To this end, this paper proposes session-based graph attention network (SGANet for short) for POI recommendation by making use of regional information. Specifically, we first extract users’ features from the regional history check-in data in session windows. Then, we use graph attention network to learn users’ preferences for both POI and regional POI, respectively. We learn the long-term and short-term preferences of users by fusing the user embedding and POI ancillary information through gate recurrent unit. Finally, we conduct experiments on two real world location-based social network datasets Foursquare and Gowalla to verify the effectiveness of the proposed recommendation model and the experiments results show that SGANet outperformed the compared baseline models in terms of recommendation accuracy, especially in sparse data and cold start scenario.

With the booming of the Internet of Things (IoT) and the speedy advancement of Location-Based Social Networks (LBSNs), Point-Of-Interest (POI) recommendation has become a vital strategy for supporting people's ability to mine their POIs. However, classical recommendation models, such as collaborative filtering, are not effective for structuring POI recommendations due to the sparseness of user check-ins. Furthermore, LBSN recommendations are distinct from other recommendation scenarios. With respect to user data, a user's check-in record sequence requires rich social and geographic information. In this paper, we propose two different neural-network models, structural deep network Graph embedding Neural-network Recommendation system (SG-NeuRec) and Deepwalk on Graph Neural-network Recommendation system (DG-NeuRec) to improve POI recommendation. combined with embedding representation from social and geographical graph information (called SG-NeuRec and DG-NeuRec). Our model naturally combines the embedding representations of social and geographical graph information with user-POI interaction representation and captures the potential user-POI interactions under the framework of the neural network. Finally, we compare the performances of these two models and analyze the reasons for their differences. Results from comprehensive experiments on two real LBSNs datasets indicate the effective performance of our model.

For DNA viruses, the immediate-early (IE) proteins are generally essential regulators that manipulate the host machinery to support viral replication. Recently, IE1, an IE protein encoded by white spot syndrome virus (WSSV), has been demonstrated to function as a transcription factor. However, the target genes of IE1 during viral infection remain poorly understood. Here, we explored the host target genes of IE1 using RNAi coupled with transcriptome sequencing analysis. A total of 429 differentially expressed genes (DEGs) were identified from penaeid shrimp, of which 284 genes were upregulated and 145 genes were downregulated after IE1 knockdown. GO and KEGG pathway enrichment analysis revealed the identified DEGs are significantly enriched in the minichromosome maintenance (MCM) complex and DNA replication, indicating that IE1 plays a critical role in DNA replication control. In addition, it was found that Penaeus vannamei MCM complex genes were remarkably upregulated after WSSV infection, while RNAi-mediated knockdown of PvMCM2 reduced the expression of viral genes and viral loads at the early infection stage. Finally, we demonstrated that overexpression of IE1 promoted the expression of MCM complex genes as well as cellular DNA synthesis in insect High-Five cells. Collectively, our current data suggest that the WSSV IE1 protein is a viral effector that modulates the host DNA replication machinery for viral replication.

BACKGROUND: The novel coronavirus disease 2019 (COVID-19) was broken out in December 2019 and soon became a global health emergency. Effective treatment for COVID-19 is urgently needed. In the present study, we aimed to evaluate the antiviral effect of Arbidol vs. Chloroquine in treating COVID-19. METHODS: We retrospectively analyzed 62 patients with COVID-19 diagnosed according to the guidelines for diagnosis and treatment of COVID-19 in China. They were divided into two groups depending on the antiviral drugs that they received. Participants in the Arbidol group (n=42) received 0.2 g Arbidol, tid for 10 days,and those in Chloroquine group (n=20) received 500 mg Chloroquine, bid for 10 days. The coronavirus negative conversion time and the length of hospital stay were analyzed and compared between the two groups. RESULTS: There was no significant difference in demographic and clinical characteristics between the two groups. After antiviral treatment, the nasopharyngeal specimen negative conversion time of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the length of hospital stay in the Arbidol group were significantly shorter than those in the Chloroquine group (18.50 vs. 25.05 days, P=0.001; 23.52 vs. 28.75 days, P=0.001). Adverse events observed during the antiviral treatment period were comparable between the two groups. Overall, 3 (7.14%) participants in the Arbidol group and 4 (20.0%) in the Chloroquine group experienced adverse events during antiviral treatment. CONCLUSIONS: These results suggest that Arbidol is advantageous over Chloroquine in terms of the SARS-CoV-2 negative conversion and the length of hospital stay in treating COVID-19 patients.

In the Notch signaling pathway in vertebrates and invertebrates, the ligand Delta plays crucial roles in cell proliferation, differentiation, and immunity. Although the Notch signaling pathway has recently been implicated in the immune defense of Penaeus vannamei, the association of Delta with this immune response remains unclear. Here, we cloned and characterized the Delta homolog in P. vannamei (designated as PvDelta). PvDelta has a 2493 bp open reading frame (ORF) encoding a putative protein of 830 amino acids. Bioinformatics analysis revealed that PvDelta contains an N-terminal signal peptide, a conserved Notch ligand (MNNL) domain, a Delta/Serrate/Lag-2 segment, 9 epidermal growth factors segments, a transmembrane domain, and shares high homology with other Delta family members. Transcripts of PvDelta were detected in all shrimp tissues tested and were induced by Vibrio parahaemolyticus, white spot syndrome virus (WSSV), and lipopolysaccharide (LPS), indicating its involvement in shrimp immune response. Moreover, after PvDelta knockdown followed by LPS stimulation, the expression of Notch signaling pathway genes (i.e., PvNotch, PvCSL, and PvHey) was downregulated. Finally, shrimp depleted of PvDelta showed a lower survival rate in response to V. parahaemolyticus challenge. In sum, our data reveal that PvDelta is involved in the innate immunity of shrimp by positively modulating the Notch signaling pathway.

In the case that user profiles are not available, the recommendation based on anonymous session is particularly important, which aims to predict the items that the user may click at the next moment based on the user's access sequence over a while. In recent years, with the development of recurrent neural network, attention mechanism, and graph neural network, the performance of session-based recommendation has been greatly improved. However, the previous methods did not comprehensively consider the context dependencies and short-term interest first of the session. Therefore, we propose a context-aware short-term interest first model (CASIF).The aim of this paper is improve the accuracy of recommendations by combining context and short-term interest. In CASIF, we dynamically construct a graph structure for session sequences and capture rich context dependencies via graph neural network (GNN), latent feature vectors are captured as inputs of the next step. Then we build the short-term interest first module, which can to capture the user's general interest from the session in the context of long-term memory, at the same time get the user's current interest from the item of the last click. In the end, the short-term and long-term interest are combined as the final interest and multiplied by the candidate vector to obtain the recommendation probability. Finally, a large number of experiments on two real-world datasets demonstrate the effectiveness of our proposed method., 15 pages,5 figures,8th International Conference on Computer Science and Information Technology (CoSIT 2021), March 27 ~ 28, 2021, Sydney, Australia

Honeydew is a watery fluid excreted by plant sap-feeding insects. It is a waste product for the insect hosts. However, it plays important roles for other organisms, such as serving as a nutritional source for beneficial insects and bacteria, as well as elicitors and effectors modulating plant responses. In this study, shotgun LC&ndash, MS/MS analyses were used to identify the proteins in the honeydew from two important rice hemipteran pests, the brown planthopper (Nilaparvata lugens, BPH) and green rice leafhopper (Nephotettix cincticeps, GRH). A total of 277 and 210 proteins annotated to insect proteins were identified in the BPH and GRH honeydews, respectively. These included saliva proteins that may have similar functions as the saliva proteins, such as calcium-binding proteins and apolipophorin, involved in rice plant defenses. Additionally, a total of 52 and 32 Oryza proteins were identified in the BPH and GRH honeydews, respectively, some of which are involved in the plant immune system, such as Pathogen-Related Protein 10, ascorbate peroxidase, thioredoxin and glutaredoxin. Coincidently, 570 and 494 bacteria proteins were identified from the BPH and GRH honeydews, respectively, which included several well-known proteins involved in the plant immune system: elongation factor Tu, flagellin, GroEL and cold-shock proteins. The results of our study indicate that the insect honeydew is a complex fluid cocktail that contains abundant proteins from insects, plants and microbes, which may be involved in the multitrophic interactions of plants&ndash, insects&ndash, microbes.

With the rapid development of location-based social networks (LBSNs), point of interest (POI) recommendation has become more and more popular personalized service. Cold start problem and poor interpretability are two main challenges of the traditional recommendation system. In this paper, we propose a novel social and sequence-aware next POI recommendation model. Our model utilizes an improved DeepWalk on heterogeneous network to learn better POI representation which contains social relationship and geographical influence. In this way, the cold start problem and interpretability problem can be solved to some extend. As for the user preference learning, we use LSTM (Long Short-Term Memory) mechanism to infer user interest, taking both long term and short term intentions into consideration. The model is trained in a metric learning framework. The experimental results on two real LBSN datasets show our model has outstanding performance in terms of AUC, Recall, and ACC. Furthermore, we can obtain 20% performance improvement on spare dataset which indicates that our model can well solve the cold start problem.

The vanilloid-type transient receptor potential (TRPV) channel is reported to be the molecular target of the commercial insecticide pymetrozine, which specifically disrupts the feeding of plant sap-sucking insects. However, the functions of TRPV channels in plant sap-sucking insects have not been fully elucidated. In the present study, RNA interference was used to investigate the effects of the knockdown of TRPV genes (Nan and Iav) on the mortality, locomotion, and feeding behavior of an important plant-feeding insect pest in rice, the brown planthopper, Nilaparvata lugens. Injecting dsRNA of Nan and Iav into fourth-instar nymphs significantly knocked down the target genes. The injection of dsNan or dsIav did not affect any morphological phenotype (including leg extension) of N. lugens nymphs and adults. Knockdown of Nan or Iav resulted in significantly decreased climbing activity against top plants but did not influence the leg-griping strength of adults. Knockdown of Nan resulted in a significantly elevated mortality of N. lugens in the observation period of 7 d after injection, whereas no significant difference in survival rates 7 d after injection was found between dsIav-injected and dsGFP-injected insects. Electropenetrographic (EPG) recordings indicated that knockdown of Nan and Iav reduced the ingestion activity in the rice phloem tissues of N. lugens. Knockdown of Nan and Iav significantly reduced the amount of honeydew excreted by N. lugens. Our findings indicated a relationship between TRPV and N. lugens locomotion and feeding behavior, which may help to fully elucidate the functions of TRPV in insects.

The problem of influence maximization (IM) in a social network is to determine a set of nodes that could maximize the spread of influence. The IM problem has been vitally applied to marketing, advertising, and public opinion monitoring. Although recent studies have studied the IM problem, they are generally greedy or heuristic-based algorithms, which are time consuming for practical use in large-scale social networks. Based on the observation that structural hole nodes usually are much more influential than other nodes, in this paper, we develop a structure-hole-based influence maximization algorithm (SHIM) with an emphasis on time efficiency. The SHIM algorithm utilizes structure hole information to significantly decrease the number of candidates of seed nodes. To measure the structure importance of nodes, we propose an structure hole value calculate algorithm to calculate the structural hole value of nodes. We prove the SHIM is NP-hard and propose a structure-based greedy algorithm to select seeds with wide influence spread and high structural hole value. We conduct experiments on real data sets to verify our algorithm’s time efficiency and accuracy, and the experimental results show that comparing with the existing algorithms, our algorithms are much more efficient and scalable.

In many applications of mobile sensor networks, such as water flow monitoring and disaster rescue, the nodes in the network can move together or separate temporarily. The dynamic network topology makes traditional spanning-tree-based aggregation algorithms invalid in mobile sensor networks. In this paper, we first present a distributed clustering algorithm which divides mobile sensor nodes into several groups, and then propose two distributed aggregation algorithms, Distance-AGG (Aggregation based on Distance), and Probability-AGG (Aggregation based on Probability). Both of these two algorithms conduct an aggregation query in three phases: query dissemination, intra-group aggregation, and inter-group aggregation. These two algorithms are efficient especially in mobile networks. We evaluate the performance of the proposed algorithms in terms of aggregation accuracy, energy efficiency, and query delay through ns-2 simulations. The results show that Distance-AGG and Probability-AGG can obtain higher accuracy with lower transmission and query delay than the existing aggregation algorithms.

Because the current research on wireless sensor networks is not thorough enough, with such shortcomings as lack of objectivity and insufficient accuracy of the trust evaluation model of network nodes, this paper proposes a trust evaluation model with multi-factor comprehensive consideration and Bayesian model, an improved trust evaluation model, by combining trust management mechanism, trust factor, fuzzy set and D-S evidence theory, solving factor singleness and imprecise evaluation in the original model. The simulation results show that the proposed model has good dynamic adaptability, higher detection rate and lower detection error rate, and it consumes less energy and improves the security of the network.

Synaptic loss underlies the memory deficit of Alzheimer's disease (AD). The molecular mechanism is elusive; however, excitatory synapses organized by the postsynaptic density (PSD) have been used as targets for AD treatment. To identify pathological entities at the synapse in AD, synaptic proteins were screened by quantitative proteomic profiling. The critical proteins were then selected for immunoblot analysis. The glutamate receptors N-methyl-d-aspartate (NMDA) receptor 1 and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor 2 (GluR2) were substantially lost; specifically, the loss of GluR2 was up to 40% at PSD in AD. Shank proteins, the organizers of these glutamate receptors at excitatory synapses, were dramatically altered in AD. The level of Shank2 was increased, whereas the protein level of Shank3 was decreased. Further, the Shank3 protein was modified by ubiquitin, indicating that abnormal activity of the ubiquitin-proteasome system may lead to Shank3 degradation in AD. Our findings suggest that disruption of glutamate receptors at the Shank-postsynaptic platform could contribute to destruction of the PSD which underlies the synaptic dysfunction and loss in AD.

Pemetrexed combined with platinum is a first-line therapy used to treat patients with advanced non-small cell lung cancer (NSCLC) that exhibit negative or unknown epidermal growth factor receptor (EGFR) mutational status or anaplastic lymphoma kinase (ALK) rearrangements. Lung adenocarcinoma (LAC) is the primary type of NSCLC. In order to prevent overtreatment, it is necessary to identify patients with LAC who may not benefit from certain chemotherapies. Patients recruited in the present study (n=129) were diagnosed with advanced LAC and received first-line pemetrexed and platinum-based chemotherapy. A microRNA (miR) microarray was used to screen the plasma miR expression profiles in a screening set of eight patients prior to and following treatment. Specifically, plasma miR-25, miR-21, miR-27b, miR-326, miR-483-5p and miR-920 were selected for reverse transcription-quantitative polymerase chain reaction analysis in a training set (n=44) prior to treatment. The screening and training set patients were all non-smokers with no prior history of serious or chronic disease. The ∆∆Cq values of these miRs were compared between the group that showed benefit from pemetrexed and platinum treatment and the group that did not. Consequently, the ∆∆Cq values of miR-25, miR-21, miR-27b and miR-326 were further determined in a validation set (n=77). The results of the present study demonstrate that plasma expression levels of miR-25, miR-21, miR-27b and miR-326, in the training and validation sets prior to treatment, were significantly different between the benefit and non-benefit groups (P≤0.001). The expression of miR-25, miR-21, miR-27b and miR-326 was upregulated in the non-benefit group and this elevation was positively correlated with decreased progression-free survival (PFS; P≤0.001). In addition, the predictive power of each miR was evaluated through receiver operating characteristic curves, in which miR-25 exhibited the highest degree of accuracy (area under the curve, 0.926; 95% confidence interval, 0.881-0.971). These results indicate that overexpression of plasma miR-25, miR-21, miR-27b and miR-326, prior to treatment, in patients with advanced LAC is predictive of non-benefit from first-line pemetrexed and platinum-based chemotherapy, and is associated with decreased PFS. Among these four miRs, miR-25 exhibited the highest degree of accuracy in predicting insensitivity, suggesting it is the most promising biomarker.

Estimating the number of distinct values in a data stream is a vital problem with many applications such as complex join query over multiple data streams. In this paper, we focus on the continuous and periodic distinct values estimation over sliding windows. We propose a compound sliding window model to compute the distinct values over basic sliding windows in an incremental way. LDV, HDV and AHDV are the three algorithms that are based on compound sliding windows. The basic idea behind the compound sliding windows is to organize the basic windows into a Hash table according to distinct values. Whenever a new data arrives at the data stream, it is inserted into a basic window. Once the basic window is full, a scan using distinct values is executed and the distinct values number is updated incrementally. Theoretical analysis and experiment results show that the distinct values estimation algorithms based on compound sliding windows have a great performance benefits.

Krüppel-like factor 8 (KLF8) regulates critical gene transcription and cellular events associated with cancer. However, KLF8-interacting proteins remain largely unidentified. Using co-immunoprecipitation (co-IP), mass spectrometry, and GST pulldown assays, we identified poly(ADP-ribose) polymerase-1 (PARP-1) as a novel KLF8-interacting protein. Co-IP and Western blotting indicated that KLF8 is also a PARP-1 substrate. Mutation of the cysteines in the zinc finger domain of KLF8 abolished PARP-1 interaction. Surprisingly, immunofluorescent staining revealed a cytoplasmic mislocalization of KLF8 in PARP-1(-/-) cells or when the interaction was disrupted. This mislocalization was prevented by either PARP-1 re-expression or inhibition of CRM1-dependent nuclear export. Interestingly, co-IP indicated competition between PARP-1 and CRM1 for KLF8 binding. Cycloheximide chase assay showed a decrease in the half-life of KLF8 protein when PARP-1 expression was suppressed or KLF8-PARP-1 interaction was disrupted. Ubiquitination assays implicated KLF8 as a target of ubiquitination that was significantly higher in PARP-1(-/-) cells. Promoter reporter assays and chromatin immunoprecipitation assays showed that KLF8 activation on the cyclin D1 promoter was markedly reduced when PARP-1 was deleted or inhibited or when KLF8-PARP-1 interaction was disrupted. Overall, this work has identified PARP-1 as a novel KLF8-binding and -regulating protein and provided new insights into the mechanisms underlying the regulation of KLF8 nuclear localization, stability, and functions.

The use of oxaliplatin is very little in the treatment of lung can- cer . This study is designed to find out the efficacy and toxicity of the combination of oxaliplatin and gemcitabine in the treatment of advanced non-small cell lung cancer (NSCLC).There were 32 cases of advanced NSCLC patients who were treated with oxaliplatin 65mg/m² by intravenous infusion on the 1st and 8th days, and gemcitabine 800-1000mg/m² by intravenous infusion on the 1st and 8th days every 21 days. After 2 cycles the efficacy was evaluated.There were 31 cases that could be evaluated. No complete response was observed and the overall response rate was 22.6% (7 partial response). The disease control rate was 71.0%. The median time to diease progression was 7 months and the median survival time was 14.5 months. The 1-year survival rate was 59.0% and 2-year survival rate was 45.8%. The main toxicities were myelosuppression and nausea and vomiting.The combination of oxaliplatin and gemcitabine is effective and well tolerated in the treatment of advanced NSCLC. The quality of life is little influenced and the regimen also can prolong the survival time in some patients.

KLF8 regulates target genes by recruiting the p300 and PCAF co-activators via glutamines (Q) 118 and 248, the CtBP co-repressor to 86PVDLS90 or SUMO to lysine (K) 67. Here we examined how these interactions coordinate to regulate KLF8 transactivity. Mass spectrometry and immunoprecipitations determined that p300 and/or PCAF promoted KLF8 acetylation at K67, K93, and K95 and this acetylation was abolished in lysine-to-arginine (R) mutants. Treatment with HDAC inhibitors or expression of co-activators inhibited sumoylation at K67. K93R or K95R mutation exerted high levels of sumoylation while the acetylation mimetic mutations K93Q and K95Q blocked the sumoylation. Interestingly, CtBP promoted sumoylation at K67 of wild-type but not AVALF mutant KLF8, and KLF8 interaction with CtBP was inhibited by treatment with the HDAC inhibitors, ectopic expression of the co-activators, or K93Q or K95Q mutation. Promoter reporter assays showed that CtBP inhibited KLF8 transactivity which was rescued by PCAF or p300 expresson. Finally, KLF8-mediated cyclin D1 protein expression and cell cycle progression were significantly decreased in the K93R and K95R but increased in the K93Q, K95Q, K67R or K67Q mutant. Taken together, these results identified a novel mechanism by which co-activators promote KLF8 transactivity by competing with SUMO for K67 modification and by acetylating K93 and K95 to inhibit CtBP-induced K67 sumoylation.