Search

Modeling users' short-term dynamic and long-term static interests to enhance Point-of-Interests (POI) recommendation performance has shown lots of advantages. Since users' check-in records can be viewed as a graph network, methods based on Graph Neural Networks (GNNs) have recently shown promising applicability for POI recommendation. However, existing GNN-based works have the following shortcomings: (1) ignoring the impact of complex higher-order relationships between user-POI dynamics over time; and (2) ignoring the difference in POI importance that cannot effectively capture the imbalances of geographical influence among POIs. To address these challenges, we propose a novel Self-supervised Long-and Short-term model (SLS-REC) for POI recommendation. Specifically, we first design a spatio-temporal Hawkes attention hypergraph neural network to capture the spatial dependence and temporal evolution in users' short-term dynamic interests. Then we introduce a dynamic propagation mechanism of GNNs to learn the geographic influences underlying geographic imbalances among POIs. In addition, the contrastive learning framework over a fine-grained node dropout strategy is applied to maximize the mutual information of long and short-term interest representations. Finally, we adaptively unify the recommendation and self-supervised task with an attention-based mechanism to optimize the proposed SLS-REC model for POI recommendation. Experiments on real-world datasets show that the proposed model significantly outperforms state-of-the-art methods. • Complex interactions between POIs are overlooked. • Hypergraph with Hawkes attention mines the dynamics of users' interests. • Novel Geographic graph model capture asymmetric geographic influence. • Fine grained contrastive learning framework fosters performance improvements. [ABSTRACT FROM AUTHOR]

Abstract Cyperus alternifolius based mesocosm-scale constructed wetland was employed to remove pharmaceuticals. We investigated the microbial community composition using phosphor lipid fatty acids (PFLAs) analysis and substrate enzyme activity during long-term exposure to pharmaceuticals in mesocosm-scale constructed wetlands. The results showed that there was no visible inhibition effect of pharmaceuticals on CW substrate enzymes activities in the experimental range (0–500 μg/L). Microbial communities, as revealed by PFLAs, were enhanced by the presence of plants, while the PFLAs content was highest when the pharmaceutical concentration was 10 μg/L or 30 μg/L at CWs. Except for anaerobic bacteria and Saturated fatty acids, the maximum PLFAs levels were reached when the pharmaceuticals were 10 μg/L or 30 μg/L, while Bacteria, G (−), fungal bacteria, Aerobic bacteria and Monounsaturated fatty acids were remarkably affected by high pharmaceuticals (100–500 μg/L). However, the main microbial florae were not changed among the treatments. In this study, the removal efficiencies of the studied pharmaceuticals in Planted (30) was greatest, which could be attributed to the higher microbial biomass. These results indicate that C. alternifolius can phytoremediate pharmaceutical-contaminated waters in CWs. Individual fatty acid cannot be used to represent specific species; therefore, more approaches to species identification such as rRNA-based methods must be included in future studies to better understand the metabolic mechanisms of microorganisms involved in the removal of studied pharmaceuticals and improve the performance of CWs. Highlights • The microbial community and enzyme activity were evaluated during long-term stress. • Enzyme activity was not inhibited in the experimental range (0–500 μg/L). • The main microbial florae were not affected in laboratory experiments. • The total PLFAs did not differ obviously in response to low pharmaceuticals in CWs. [ABSTRACT FROM AUTHOR]

Soluble-free, soluble-conjugated, insoluble-bound phenolics and antioxidant activity, flavonoid (TFC), proanthocyanidins (TPAC), anthocyanins and minerals of fifteen whole rice grains with different colors were investigated. Soluble-free protocatechuic and vanillic acids were only quantified in black rice, which had the most quantities. Non-pigmented rice had no detectable conjugated protocatechuic and 2,5 -dihydroxybenzoic acids both of which were found in black and red rice, respectively. The main bound phenolic acids were ferulic and p -coumaric, as well as 2,5 -dihydroxybenzoic in red rice and protocatechuic and vanillic acids in black rice. Soluble-conjugated phenolics, TFC, and anthocyanins were negatively correlated with L ∗ , b ∗ , C and H ° values. TPAC was positively correlated with a ∗ ( P < 0.01). Protocatechuic, vanillic, syringic and ferulic acids were associated with TPC and antioxidant activity in the soluble-conjugated fraction while protocatechuic and ferulic acid were correlated with those in the insoluble-bound fraction. Principal component analysis divided samples into non-pigmented, red and black rice groups. [ABSTRACT FROM AUTHOR]

The presence of pharmaceutically active compounds (PhACs) in wastewater treatment plant effluent poses a potential risk to aquatic ecosystems. Constructed wetlands have recently been used to control PhACs. However, the microbial communities that are involved in these processes have not been comprehensively investigated. This study aimed to evaluate the removal of PhACs and microbial response in constructed wetlands during the treatment of PhACs. The effects of PhACs on bacterial communities in constructed wetland mesocosms were analyzed by Illumina MiSeq sequencing technology. Results indicated that removal efficiencies of PhACs were enhanced over time, and constructed wetlands offer higher removal efficiencies for the PhACs studied compared to conventional wastewater treatment plants. Plants improved microbial richness and diversity while both indices were negatively correlated with PhAC concentrations ranging from 30 to 500 μg/L in constructed wetland mesocosms. The microbial communities of the constructed wetland mesocosms were dominated by Proteobacteria, Acidobacteria, and Bacteroidetes under PhAC exposure, while Desulfobulbus and Treponema were the dominant genera. In particular, Proteobacteria were correlated with PhAC concentrations. Overall, this study provides valuable microbial community ecology data to understand how microbial populations respond to PhAC stress in constructed wetlands. [ABSTRACT FROM AUTHOR]

The genotype of rice variety is important factor that affect grain quality during storage. In this study, japonica , indica , and indica – japonica hybrid rice were selected to reveal the changes in amylase activities, starch fine structure, and physicochemical properties after one year of storage. Amylase activities decreased after storage and starch molecules were degraded, which, respectively, decreased and increased the content of the amylopectin long chains (DP 37+) and amylose. Changes in the starch fine structure resulted in lower swelling factors and a higher gelatinization temperature. Compared with japonica and indica–japonica hybrid rice, indica rice had a more stable starch fine structure and physicochemical properties after storage. The information obtained from this study provides a better understanding of the changes that occur in starch at the molecular level during storage and will provide insight into the breeding of storable rice varieties. • Amylose short chains increased after storage. • Amylopectin long chains degraded after storage. • Indica rice showed fewer changes in starch molecular structure after storage. [ABSTRACT FROM AUTHOR]

A new selectable marker system was adapted for use in Agrobacterium-mediated transformation of rice (Oryza sativa L.). This selection system utilizes the pmi gene encoding for phosphomannose-isomerase that converts mannose-6-phosphate to fructose-6-phosphate. Only transformed cells were capable of utilizing mannose as a carbon source. Transgenic rice plants regenerated from selected transformed immature embryo-derived calli on media containing various concentrations of mannose. The highest transformation frequency of 6.0% was obtained when a combination of 25 g/l mannose and 5 g/l sucrose was used. Molecular and genetic analyses showed that the plants contained the pmi gene and the gene was transmitted to the progeny in Mendelian fashion. The results indicated that the mannose selection system, which was devoid of the disadvantages of antibiotic or herbicide selection, could be used for rice Agrobacterium-mediated immature embryo transformation. [Copyright &y& Elsevier]