CausalGNN: Causal-Based Graph Neural Networks for Spatio-Temporal Epidemic Forecasting

Infectious disease forecasting has been a key focus in the recent past owing to the COVID-19 pandemic and has proved to be an important tool in controlling the pandemic. With the advent of reliable spatiotemporal data, graph neural network models have been able to successfully model the interrelation between the cross-region signals to produce quality forecasts, but like most deep-learning models they do not explicitly incorporate the underlying causal mechanisms. In this work, we employ a causal mechanistic model to guide the learning of the graph embeddings and propose a novel learning framework - Causal-based Graph Neural Network (CausalGNN) that learns spatiotemporal embedding in a latent space where graph input features and epidemiological context are combined via a mutually learning mechanism using graph-based non-linear transformations. We design an attention-based dynamic GNN module to capture spatial and temporal disease dynamics. A causal module is added to the framework to provide epidemiological context for node embedding via ordinary differential equations. Extensive experiment son forecasting daily new cases of COVID-19 at global, US state, and US county levels show that the proposed method outperforms a broad range of baselines. The learned model which incorporates epidemiological context organizes the embedding in an efficient way by keeping the parameter size small leading to robust and accurate forecasting performance across various datasets.

[Consideration on the possible etiological mechanisms and countermeasures about severe acute hepatitis of unknown origin in children].

Since April 2022, severe acute hepatitis of unknown origin in children has spread to 35 countries and regions around the world, and more than 1 010 cases have been reported. Since the severe acute hepatitis of unknown origin involves a wide range of areas and has a high rate, it is critical to identify the etiology and establish effective preventive, diagnostic and therapeutic measures as soon as possible. This study discusses the possible mechanisms and countermeasures of the severe acute hepatitis of unknown origin in children. It speculates that the occurrence of the recent severe acute hepatitis might be related to adenovirus, adeno-associated virus infection, and the COVID-19 epidemic, while the difference in HLA polymorphism among different races might be related to the fact that reported cases were more common in Europe and the United States. Based on the currently available evidence, it can be preliminarily judged that the risk of large-scale outbreak of severe acute hepatitis of unknown origin in children would be low in China, but the persistent awareness and vigilance of the etiology is still needed.

Integrating network pharmacology and molecular docking to explore the potential mechanism of Xinguan No. 3 in the treatment of COVID-19

Xinguan No. 3 has been recommended for the treatment of coronavirus disease 2019 (COVID-19);however, its potential mechanisms are unclear. This study aims to explore the mechanisms of Xinguan No. 3 against COVID-19 through network pharmacology and molecular docking. We first searched the ingredients of Xinguan No. 3 in three databases (Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform, Traditional Chinese Medicines Integrated Database, and The Encyclopedia of Traditional Chinese Medicine). The active components and their potential targets were predicted through the SwissTargetPrediction website. The targets of COVID-19 can be found on the GeneCards website. Protein interaction analysis, screening of key targets, functional enrichment of key target genes, and signaling pathway analysis were performed through Search Tool for the Retrieval of Interacting Genes databases, Metascape databases, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway databases. Finally, the affinity of the key active components with the core targets was verified by molecular docking. The results showed that five core targets had been screened, including MAPK1, NF-kappa B1, RELA, AKT1, and MAPK14. Gene ontology enrichment analysis revealed that the key targets were associated with inflammatory responses and responses to external stimuli. KEGG enrichment analysis indicated that the main pathways were influenza A, hepatitis B, Toll-like receptor signaling pathway, NOD-like receptor signaling pathway, and TNF signaling pathway. Therefore, Xinguan No. 3 might play a role in treating COVID-19 through anti-inflammatory, immune responses, and regulatory responses to external stimuli.

Effects of Country and Geopolitical Risks on Income Inequality: Evidence from Emerging Economies

Income inequality is rising due to the risks and uncertainties related to the COVID-19 pandemic and other risks. This paper examines the effects of country risks (measured by economic/financial and political risks) and geopolitical risks on the income inequality in the panel dataset of 19 emerging market economies from 1985 to 2020. It is observed that all risk measures are positively related to income inequality. This evidence is also valid when different empirical models and estimation procedures are considered. The results are also robust for including various controls, excluding the extreme observations in the dataset, and considering the countries at the different income levels and regions.

FedCM: A Real-time Contribution Measurement Method for Participants in Federated Learning

Federated Learning (FL) creates an ecosystem for multiple agents to collaborate on building models with data privacy consideration. The method for contribution measurement of each agent in the FL system is critical for fair credits allocation but few are proposed. In this paper, we develop a real-time contribution measurement method FedCM that is simple but powerful. The method defines the impact of each agent, comprehensively considers the current round and the previous round to obtain the contribution rate of each agent with attention aggregation. Moreover, FedCM updates contribution every round, which enable it to perform in real-time. Real-time is not considered by the existing approaches, but it is critical for FL systems to allocate computing power, communication resources, etc. Compared to the state-of-the-art method, the experimental results show that FedCM is more sensitive to data quantity and data quality under the premise of real-time. Furthermore, we developed federated learning open-source software based on FedCM. The software has been applied to identify COVID-19 based on medical images.

A fuzzy comprehensive evaluation of climate change on the Xiamen tourism industry

Purpose - Climate change is most apparent through the increased severity and frequency of extreme events. Tourism as an activity is particularly sensitive. This paper aims to investigate the impact that climate change has on Xiamen tourism through a fuzzy comprehensive evaluation of questionnaire responses. Design/methodology/approach - A fuzzy classification system of tourism factors most sensitive to climate change was built on the basis of an analytical hierarchical process. Findings A - "relatively strong" association grade of the impacts of climate change on tourism was observed. Through fuzzy comprehensive evaluation, the method used has allowed for clear classification of the aspects of tourism, through its development, which are more vulnerable to climate change. The results acquired here can serve as reference material for stakeholders on implementing risk assessments, deepening the understanding of how climate change affects tourism and coordinate the interests of different parties through the achievement of focused development and realize the optimum, long-term and sustainable exploitation of tourism resources. Originality/value - The sensitivity of a variety of tourist sectors within Xiamen was assessed and represents the newest pre-COVID-19 opinions concerning the effect of climate change on tourism. Additionally, the data used in this study was also collected before the outbreak of the COVID-19 pandemic and will serve as an important marker to track how expert opinions of the effects of climate change on tourism change over time.

Big Data Analysis for Effects of the COVID-19 Outbreak on Ambient PM2.5 in Areas that Were Not Locked Down

COVID-19, which is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), first broke out at the end of 2019. Despite rapidly spreading around the world during the first half of 2020, it remained well controlled in Taiwan without the implementation of a nationwide lockdown. This study aimed to evaluate the PM2.5 concentrations in this country during the 2020 COVID-19 pandemic and compare them with those during the corresponding period from 2019. We obtained measurements (taken every minute or every 3 minutes) from approximately 1,500 PM2.5 sensors deployed in industrial areas of northern and southern Taiwan for the first quarters (January-March) of both years. Our big data analysis revealed that the median hourly PM2.5 levels decreased by 3.70% (from 16.3 to 15.7 mu g m(-3)) and 10.6% (from 32.4 to 29.3 mu g m(-3)) in the north and south, respectively, between these periods owing to lower domestic emissions of PM2.5 precursors (viz., nitrogen dioxide and sulfur dioxide) and, to a lesser degree, smaller transported emissions of PM2.5, e.g., from China. Additionally, the spatial patterns of the PM2.5 in both northern and southern Taiwan during 2020 resembled those from the previous year. Finally, controlling local PM2.5 emission sources critically contributes to reducing the number of COVID-19 cases.

High expression of ACE2 in the human lung leads to the release of IL6 by suppressing cellular immunity: IL6 plays a key role in COVID-19.

OBJECTIVE: The pathogenesis of coronavirus disease 2019 (COVID-19) remains clear, and no effective treatment exists. SARS-CoV-2 is the virus that causes COVID-19 and uses ACE2 as a cell receptor to invade human cells. Therefore, ACE2 is a key factor to analyze the SARS-CoV-2 infection mechanism. MATERIALS AND METHODS: We included 9,783 sequencing results of different organs, analyzed the effects of different ACE2 expression patterns in organs and immune regulation. RESULTS: We found that ACE2 expression was significantly increased in the lungs and digestive tract. The cellular immunity of individuals with elevated ACE2 expression is activated, whereas humoral immunity is dampened, leading to the release of many inflammatory factors dominated by IL6. Furthermore, by studying the sequencing results of SARS-CoV-2-infected and uninfected cells, IL6 was found to be an indicator of a significant increase in the number of infected cells. However, although patients with high expression of ACE2 will release many inflammatory factors dominated by IL6, cellular immunity in the colorectum is significantly activated. This effect may explain why individuals with SARS-CoV-2 infection have severe lung symptoms and digestion issues, which are important causes of milder symptoms. CONCLUSIONS: This finding indicates that ACE2 and IL6 inhibitors have important value in COVID-19.