Soft Sensor with Deep Learning for Functional Region Detection in Urban Environments.

The rapid development of urbanization has increased traffic pressure and made the identification of urban functional regions a popular research topic. Some studies have used point of interest (POI) data and smart card data (SCD) to conduct subway station classifications; however, the unity of both the model and the dataset limits the prediction results. This paper not only uses SCD and POI data, but also adds Online to Offline (OTO) e-commerce platform data, an application that provides customers with information about different businesses, like the location, the score, the comments, and so on. In this paper, these data are combined to and used to analyze each subway station, considering the diversity of data, and obtain a passenger flow feature map of different stations, the number of different types of POIs within 800 m, and the situation of surrounding OTO stores. This paper proposes a two-stage framework, to identify the functional region of subway stations. In the passenger flow stage, the SCD feature is extracted and converted to a feature map, and a ResNet model is used to get the output of stage 1. In the built environment stage, the POI and OTO features are extracted, and a deep neural network with stacked autoencoders (SAE-DNN) model is used to get the output of stage 2. Finally, the outputs of the two stages are connected and a SoftMax function is used to make the final identification of functional region. We performed experimental testing, and our experimental results show that the framework exhibits good performance and has a certain reference value in the planning of subway stations and their surroundings, contributing to the construction of smart cities.

Forecasting Subway Passenger Flow for Station-Level Service Supply.

Demand forecasting is one of the managers' concerns in service supply chain management. With accurate passenger flow forecasting, the station-level service suppliers can make better service plans accordingly. However, the existing forecasting model cannot identify the different future passenger flow at different types of stations. As a result, the service suppliers cannot make service plans according to the demands of different stations. In this article, we propose a deep learning architecture called DeepSPF (Deep Learning for Subway Passenger Forecasting) to predict subway passenger flow considering the different functional types of stations. We also propose the sliding long short-term memory (LSTM) neural networks as an important component of our model, combining LSTM and one-dimensional convolution. In the experiments of the Beijing subway, DeepSPF outperforms the baseline models in three-time granularities (10, 15, and 30 minutes). Moreover, a comparison between variants of DeepSPF indicates that, with the information of stations' functional types, DeepSPF has strong robustness when an abnormal situation happens.

Identification of porcine serum proteins modified in response to HP-PRRSV HuN4 infection by two-dimensional differential gel electrophoresis.

Since 2006, highly pathogenic porcine reproductive and respiratory syndrome virus (HP-PRRSV) has become the major pathogen attributed to the prevalent porcine reproductive and respiratory syndrome (PRRS) in China. The present study aims to identify serum proteins modified in response to infection of HuN4, a HP-PRRSV strain isolated from a farm in 2006. 2-D DIGE analysis allowed for the detection of 19 differentially expressed protein spots, of which 18 were identified by MALDI-TOF/TOF MS. These 18 spots represented for a total of 9 proteins (6 up-regulated and 3 down-regulated), most of which belonged to the acute phase proteins in swine and showed a trend of regression in the late phase of the experiment. One of a series of AGP spots was identified for the first time to be decreased in acute phase of PRRSV infection in swine. But the whole level of the protein in the serum did not show significant changes by Western blot. The rising tendency of Hp was confirmed by Western blot and ELISA. These altered proteins were probably involved in the inflammatory process triggered by HuN4 and in alleviating the oxidative damage occurring in the process. In summary, these results may provide new insights into understanding the mechanisms of HP-PRRSV infection.

Highly pathogenic porcine reproductive and respiratory syndrome virus GP5 B antigenic region is not a neutralizing antigenic region.

In 2006, highly pathogenic porcine reproductive and respiratory syndrome virus (HP-PRRSV) caused great economic losses emerged in China and continues to be a threat for the pig industry. B antigenic region (AR) ((37)SHL/FQLIYNL(45)) of GP5 was considered to be a major linear neutralizing AR in PRRSV classical strains. However, peptide-purified antibodies against this AR did not neutralize PRRSV in a recent report. Compared with classical PRRSV, one amino acid mutation (L/F(39)→ I(39)) was found in B AR of HP-PRRSV. To study the ability of B AR of HP-PRRSV to induce neutralizing antibody (NA) in vitro and in vivo, rabbit antisera against B AR with and without the mutation and pig hyperimmune sera with high titer of NAs against HP-PRRSV were prepared. Immunofluorescence assays (IFA) showed that the two rabbit antisera both had reactivity to classical PRRSV CH-1a and HP-PRRSV HuN4 with no observable difference in IFA titer. However, antisera did not have neutralizing activity against classical PRRSV CH-1a and HP-PRRSV HuN4. No correlation was observed between the levels of anti-B AR peptide antibodies and NAs in pig hyperimmune sera that were detected by indirect ELISA and virus neutralization, respectively. B AR peptide-specific serum antibodies had no neutralizing activity and, GST-B fusion protein could not inhibit neutralization of NAs in pig hyperimmune sera. Based on these findings, we conclude that B AR of HP-PRRSV is not a neutralizing AR of HP-PRRSV GP5.

Isolation of serotype 2 porcine teschovirus in China: evidence of natural recombination.

Porcine teschovirus (PTV), the pathogen of porcine polioencephalomyelitis, is a member of the family Picornaviridae. In this study, a new PTV strain (designated as JF613) was isolated from pigs in China. It was confirmed by the specific CPE on susceptible cells, RT-PCR and nucleotide sequencing. Analysis of its amino acids sequence of complete polyprotein indicated that the isolate belongs to serotype 2. Genetic recombination is a well-known phenomenon for picornavirus which has been demonstrated in many other members of the family, but it remains so far unclear whether recombination occurs in PTV. To detect possible recombination events, 30 sequences of complete coding regions of PTV strains accessible in GenBank were examined. Putative recombinant sequence was identified with the use of SimPlot program. The result showed that the genomic sequence of our isolate exhibited highest similarities with strains of serotypes 2 and 5, respectively, in two crossover regions, suggesting the recombination event in PTV. Then the mosaic structure of viral genome was confirmed by bootscanning and genetic algorithm for recombination detection (GARD). This represents the first PTV-2 isolate in China. Furthermore, our study provided the first evidence of natural recombination in PTV and indicated that homologous recombination may be a driving force in PTV evolution.