Immune and regulative characterization of complement-related gene cfhl5 in response to Vibrio harveyi challenge in Cynoglossus semilaevis.

Complement factor H-related protein (CFHR) plays an important role in regulating complement activation and defensive responses. The function of CFHR2 (complement factor H related 2), a member of the CFHR family, in fish remains unclear. Here, we report the genetic relationship, expression characteristics and regulatory mechanism of cfhl5 (complement factor H like 5) gene, which encodes CFHR2 in Chinese tongue sole. We observed that the cfhl5 gene was widely expressed in several tissues, such as brain, heart and immune organs, and was most abundantly expressed in liver. After injection with Vibrio harveyi, the expression of cfhl5 was up-regulated significantly in liver, spleen and kidney at 12 or 24 hours post infection (hpi), suggesting an involvement of this gene in the acute immune response. Knockdown of cfhl5 in liver cells significantly up-regulated the expression of the pro-inflammatory cytokines tnf-α (tumor necrosis factor-alpha) and il1ß (interleukin-1beta), the immunomodulatory factor il10 (interleukin-10) and the lectin complement pathway gene masp1 (MBL-associated serine protease 1), and down-regulated the expression of complement components c3 (complement 3) and cfi (complement factor I). In our previous work, we found that cfhl5 gene was significantly higher methylated and lower expressed in the resistant family compared with the susceptible family. Therefore, we used dual-luciferase reporter system to determine the effect of DNA methylation on this gene and found that DNA methylation could inhibit the promoter activity to reduce its expression. These results demonstrated that the expression of cfhl5 is regulated by DNA methylation, and this gene might play an important role in the immune response by regulating the expression of cytokines and complement components genes in Chinese tongue sole.

An adaptive adjacency matrix-based graph convolutional recurrent network for air quality prediction.

In recent years, air pollution has become increasingly serious and poses a great threat to human health. Timely and accurate air quality prediction is crucial for air pollution early warning and control. Although data-driven air quality prediction methods are promising, there are still challenges in studying spatial-temporal correlations of air pollutants to design effective predictors. To address this issue, a novel model called adaptive adjacency matrix-based graph convolutional recurrent network (AAMGCRN) is proposed in this study. The model inputs Point of Interest (POI) data and meteorological data into a fully connected neural network to learn the weights of the adjacency matrix thereby constructing the self-ringing adjacency matrix and passes the pollutant data with this matrix as input to the Graph Convolutional Network (GCN) unit. Then, the GCN unit is embedded into LSTM units to learn spatio-temporal dependencies. Furthermore, temporal features are extracted using Long Short-Term Memory network (LSTM). Finally, the outputs of these two components are merged and air quality predictions are generated through a hidden layer. To evaluate the performance of the model, we conducted multi-step predictions for the hourly concentration of PM2.5, PM10 and O3 at Fangshan, Tiantan and Dongsi monitoring stations in Beijing. The experimental results show that our method achieves better predicted effects compared with other baseline models based on deep learning. In general, we designed a novel air quality prediction method and effectively addressed the shortcomings of existing studies in learning the spatio-temporal correlations of air pollutants. This method can provide more accurate air quality predictions and is expected to provide support for public health protection and government environmental decision-making.

Genome-wide DNA methylation mediates the resistance to vibriosis in Cynoglossus semilaevis.

Chinese tongue sole (Cynoglossus semilaevis) is an economically important marine fish in China. However, vibriosis has caused huge mortality and economic losses in its culturing industry. To reveal the effect of DNA methylation on the resistance to vibriosis in tongue sole, we conducted RNA sequencing and whole genome bisulfite sequencing (WGBS), and compared the gene expressions and DNA methylation patterns between the resistant and susceptible families. We identified a total of 741 significantly differentially expressed genes (DEGs) in kidney and 17460 differentially methylated genes (DMGs), which were both enriched in immune-related pathways, such as "cAMP signaling pathway" and "NOD-like receptor signaling pathway". Through the correlation analysis of DEGs and DMGs, we identified two important immune pathways, including "complement and coagulation cascades", and "cytokine-cytokine receptor interaction", which played important roles in regulating the inflammation level and immune homeostasis. For example, the expression of proinflammatory cytokine il17c was down-regulated under the regulation of DNA methylation; in addition, the expression of protease-activated receptor 3 (par3) was up-regulated, which could induce the up-expressionof il8. These results demonstrated that the regulation of DNA methylation on the genes involved in immune responses might contribute to the resistance to vibriosis in tongue sole, and provided a basis for the control of diseases in fish aquaculture.

Correction: Empirical mode decomposition based long short-term memory neural network forecasting model for the short-term metro passenger flow.

[This corrects the article DOI: 10.1371/journal.pone.0222365.].

Empirical mode decomposition based long short-term memory neural network forecasting model for the short-term metro passenger flow.

Short-term metro passenger flow forecasting is an essential component of intelligent transportation systems (ITS) and can be applied to optimize the passenger flow organization of a station and offer data support for metro passenger flow early warning and system management. LSTM neural networks have recently achieved remarkable recent in the field of natural language processing (NLP) because they are well suited for learning from experience to predict time series. For this purpose, we propose an empirical mode decomposition (EMD)-based long short-term memory (LSTM) neural network model for predicting short-term metro inbound passenger flow. The EMD algorithm decomposes the original sequential passenger flow into several intrinsic mode functions (IMFs) and a residual. Selected IMFs that are strongly correlated with the original data can be obtained via feature selection. The selected IMFs and the original data are integrated into inputs for LSTM neural networks, and a single LSTM prediction model and an EMD-LSTM hybrid forecasting model are developed. Finally, historical real automatic fare collection (AFC) data from metro passengers are collected from Chengdu Metro to verify the validity of the proposed EMD-LSTM prediction model. The results indicate that the proposed EMD-LSTM hybrid forecasting model outperforms the LSTM, ARIMA and BPN models.