This paper investigates the problem of bootstrapping a statistical dialogue manager without access to training data and proposes a new probabilistic agenda-based method for simulating user behaviour. In experiments with a statistical POMDP dialogue system, the simulator was realistic enough to successfully test the prototype system and train a dialogue policy. An extensive study with human subjects showed that the learned policy was highly competitive, with task completion rates above 90%.
Abstract-The seasonal variations of estrogenic compounds and the estrogenicities of influent and effluent were investigated by chemical analysis and in vitro assay in a municipal sewage treatment plant in Wuhan (China). The levels of eight estrogenic compounds, including 17-estradiol (E 2 ), estrone (E 1 ), estriol (E 3 ), diethylstilbestrol (DES), 17␣-ethinylestradiol, nonylphenol (NP), 4-tert-octylphenol (OP), and bisphenol A (BPA), were measured by gas chromatography-mass spectrometry. Total estrogenic activity of sewage was quantitatively assessed using primary cultured hepatocytes of male Megalobrama amblycephala Yih using vitellogenin as a biomarker. The E 2 equivalents (EEQs) obtained from the chemical analysis were consistent with those measured by bioassay. The natural (E 1 , E 2 , and E 3 ) and synthetic (DES) estrogens, as well as NP, were the main contributors of the total EEQs of influent and effluent in the present study. The levels of natural estrogens E 1 and E 3 in the influent and effluent were higher in winter than in summer, whereas the situation for NP and OP was the reverse. The levels of E 2 , DES, and BPA varied little among different seasons. 17␣-Ethinylestradiol was not detected in the influent and effluent. The estrogenicities of the influent and of the primary and secondary effluents were all higher in summer than in winter. Estrogenic activities in winter mainly originated from natural (E 1 , E 2 , and E 3 ) and synthetic (DES) estrogens, whereas the increase of EEQs in summer was contributed by NP. The results from chemical analysis and bioassay demonstrate that estrogenic compounds cannot be entirely removed by the existing sewage treatment process, which should be further improved to protect aquatic ecosystems and human health.
Due to limited morphological difference, the two closely related sister species, the cotton bollworm, Helicoverpa armigera (Hübner) and the oriental tobacco budworm, H. assulta (Guenée) (Lepidoptera: Noctuidae), are very difficult to distinguish, especially at the larvae stage. Recently, DNA sequence has been widely used as a bio-barcode for species identification. In this study, we attempted to distinguish H. armigera and H. assulta using the mitochondrial cytochrome C oxidase subunit I gene (COI) gene sequence as the barcode. We determined a 658 bp segment of the COI gene for 28 individuals of H. armigera, 8 individuals of H. assulta, and 10 individuals of Mamestra brassicae (as the outgroup) in Yunnan Province, southwest of P. R. China, together with one H. assulta and two H. armigera reported sequences from GenBank. Twenty-three haplotypes were identified in all 49 samples. As expected, network analysis of the haplotypes of the three species presented a clustering pattern consistent with the respective species status. Haplotypes of the same species differed from each other by no more than three nucleotide substitutions. However, each haplotype of H. armigera differed from that of H. assulta by at least 22 nucleotide substitutions. Both species differed from M. brassicae by more than 50 nucleotide substitutions. 17 unique diagnostic nucleotides were also used to discriminate the two species. The finding of large COI sequence differences between H. armigera and H. assulta suggested that it could be used to distinguish the two morphologically alike species and be employed for quick species identification during pest control.
This study investigated the distribution of antibiotic resistant Escherichia coli (E. coli) and examined the possible relationship between water quality parameters and antibiotic resistance from two different drinking water sources (the Qiantang River and the Dongtiao Stream) in Hangzhou city of China. E. coli isolates were tested for their susceptibility to 18 antibiotics. Most of the isolates were resistant to tetracycline (TE), followed by ampicillin (AM), piperacillin (PIP), trimethoprim/sulfamethoxazole (SXT), and chloramphenicol (C). The antibiotic resistance rate of E. coli isolates from two water sources was similar; For E. coli isolates from the Qiantang River, their antibiotic resistance rates decreased from up- to downstream. Seasonally, the dry and wet season had little impact on antibiotic resistance. Spearman's rank correlation revealed significant correlation between resistance to TE and phenicols or ciprofloxacin (CIP), as well as quinolones (ciprofloxacin and levofloxacin) and cephalosporins or gentamicin (GM). Pearson's chi-square tests found certain water parameters such as nutrient concentration were strongly associated with resistance to some of the antibiotics. In addition, tet genes were detected from all 82 TE-resistant E. coli isolates, and most of the isolates (81.87%) contained multiple tet genes, which displayed 14 different combinations. Collectively, this study provided baseline data on antibiotic resistance of drinking water sources in Hangzhou city, which indicates drinking water sources could be the reservoir of antibiotic resistance, potentially presenting a public health risk.
Sydowia polyspora occurs often as an epiphyte or endophyte of conifers and is widely distributed around the world. It is also reported as a fungus associated with bark beetles. In this study, based on ITS sequence data and morphological characteristics, 14 strains collected from the beetle bodies and galleries of Tomicus minor and Tomicus yunnanensis infesting Pinus yunnanensis in Southwestern China were identified as S. polyspora. To assess the virulence of S. polyspora to P. yunnanensis, pathogenicity tests were conducted by inoculating the fungus in the stems and on needles of P. yunnanensis. The results after 30 days showed that the lesion lengths in the phloem after inoculation with S. polyspora were significantly longer than for the control. In addition, the fungus caused partly distinct chlorisis or discolouration of needles 14 days after inoculation. This suggests that S. polyspora is able to colonize stem phloem and pine needles of P. yunnanensis and thus may cause secondary damages to this host after being spread with bark beetles.
RNAi is an effective tool for gene function analysis and a promising strategy to provide environmentally friendly control approaches for pathogens and pests. Recent studies support the utility of bacterium-mediated RNAi as a cost-effective method for gene function study and a suitable externally applied delivery mechanism for pest control. Here, we developed a bacterium-mediated RNAi system in Spodoptera frugiperda based on four target genes, specifically, Chitinase (Sf-CHI), Chitin synthase B (Sf-CHSB), Sugar transporter SWEET1 (Sf-ST), and Hemolin (Sf-HEM). RNAi conducted by feeding larvae with bacteria expressing dsRNAs of target genes or injecting pupae and adults with bacterially synthesized dsRNA induced silencing of target genes and resulted in significant negative effects on growth and survival of S. frugiperda. However, RNAi efficiency and effects were variable among different target genes and dsRNA delivery methods. Injection of pupae with dsCHI and dsCHSB induced a significant increase in wing malformation in adults, suggesting that precise regulation of chitin digestion and synthesis is crucial during wing formation. Injection of female moths with dsHEM resulted in lower mating, fecundity, and egg hatching, signifying a critical role of Sf-HEM in the process of egg production and/or embryo development. Our collective results demonstrate that bacterium-mediated RNAi presents an alternative technique for gene function study in S. frugiperda and a potentially effective strategy for control of this pest, and that Sf-CHI, Sf-CHSB, Sf-ST, and Sf-HEM encoding genes can be potent targets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.