A novel bat coronavirus with a polybasic furin-like cleavage site.

The current pandemic of COVID-19 caused by a novel coronavirus, severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), threatens human health around the world. Of particular concern is that bats are recognized as one of the most potential natural hosts of SARS-CoV-2; however, coronavirus ecology in bats is still nascent. Here, we performed a degenerate primer screening and next-generation sequencing analysis of 112 bats, collected from Hainan Province, China. Three coronaviruses, namely bat betacoronavirus (Bat CoV) CD35, Bat CoV CD36 and bat alphacoronavirus CD30 were identified. Bat CoV CD35 genome had 99.5% identity with Bat CoV CD36, both sharing the highest nucleotide identity with Bat Hp-betacoronavirus Zhejiang2013 (71.4%), followed by SARS-CoV-2 (54.0%). Phylogenetic analysis indicated that Bat CoV CD35 formed a distinct clade, and together with Bat Hp-betacoronavirus Zhejiang2013, was basal to the lineage of SARS-CoV-1 and SARS-CoV-2. Notably, Bat CoV CD35 harbored a canonical furin-like S1/S2 cleavage site that resembles the corresponding sites of SARS-CoV-2. The furin cleavage sites between CD35 and CD36 are identical. In addition, the receptor-binding domain of Bat CoV CD35 showed a highly similar structure to that of SARS-CoV-1 and SARS-CoV-2, especially in one binding loop. In conclusion, this study deepens our understanding of the diversity of coronaviruses and provides clues about the natural origin of the furin cleavage site of SARS-CoV-2.

Improving the resident assessment process: application of App-based e-training platform and lean thinking.

BACKGROUND: The assessment system for standardized resident training is crucial for developing competent doctors. However, it is complex, making it difficult to manage. The COVID-19 pandemic has also aggravated the difficulty of assessment. We, therefore, integrated lean thinking with App-based e-training platform to improve the assessment process through Define-Measure-Analyze-Improve-Control (DMAIC) cycles. This was designed to avoid unnecessary activities that generate waste. METHODS: Panels and online surveys were conducted in 2021-2022 to find the main issues that affect resident assessment and the root causes under the frame of waste. An online app was developed. Activities within the process were improved by brainstorming. Online surveys were used to improve the issues, satisfaction, and time spent on assessment using the app. RESULTS: A total of 290 clinical educators in 36 departments responded to the survey, and 153 clinical educators used the online app for assessment. Unplanned delay or cancellation was defined as the main issue. Eleven leading causes accounted for 87.5% of the issues. These were examiner time conflict, student time conflict, insufficient examiners, supervisor time conflict, grade statistics, insufficient exam assistants, reporting results, material archiving, unfamiliarity with the process, uncooperative patients, and feedback. The median rate of unplanned delay or cancellation was lower with use of the app (5% vs 0%, P < 0.001), and satisfaction increased (P < 0.001). The median time saved by the app across the whole assessment process was 60 (interquartile range 60-120) minutes. CONCLUSIONS: Lean thinking integrated with an App-based e-training platform could optimize the process of resident assessment. This could reduce waste and promote teaching and learning in medical education.

Characteristics of patients with SARS-COV-2 PCR re-positivity after recovering from COVID-19.

The purpose of this study was to analyse the clinical characteristics of patients with severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) PCR re-positivity after recovering from coronavirus disease 2019 (COVID-19). Patients (n = 1391) from Guangzhou, China, who had recovered from COVID-19 were recruited between 7 September 2021 and 11 March 2022. Data on epidemiology, symptoms, laboratory test results and treatment were analysed. In this study, 42.7% of recovered patients had re-positive result. Most re-positive patients were asymptomatic, did not have severe comorbidities, and were not contagious. The re-positivity rate was 39%, 46%, 11% and 25% in patients who had received inactivated, mRNA, adenovirus vector and recombinant subunit vaccines, respectively. Seven independent risk factors for testing re-positive were identified, and a predictive model was constructed using these variables. The predictors of re-positivity were COVID-19 vaccination status, previous SARs-CoV-12 infection prior to the most recent episode, renal function, SARS-CoV-2 IgG and IgM antibody levels and white blood cell count. The predictive model could benefit the control of the spread of COVID-19.

The Threat of Potentially Pathogenic Bacteria in the Feces of Bats.

Bats have attracted global attention because of their zoonotic association with severe acute respiratory syndrome coronavirus (SARS-CoV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Previous and ongoing studies have predominantly focused on bat-borne viruses; however, the prevalence or abundance of bat-borne pathogenic bacteria and their potential public health significance have largely been neglected. For the first time, this study used both metataxonomics (16S rRNA marker gene sequencing) and culturomics (traditional culture methods) to systematically evaluate the potential public health significance of bat fecal pathogenic bacteria. To this end, fecal samples were obtained from five bat species across different locations in China, and their microbiota composition was analyzed. The results revealed that the bat microbiome was most commonly dominated by Proteobacteria, while the strictly anaerobic phylum Bacteroidetes occupied 35.3% of the relative abundance in Rousettus spp. and 36.3% in Hipposideros spp., but less than 2.7% in the other three bat species (Taphozous spp., Rhinolophus spp., and Myotis spp.). We detected 480 species-level phylotypes (SLPs) with PacBio sequencing, including 89 known species, 330 potentially new species, and 61 potentially higher taxa. In addition, a total of 325 species were identified by culturomics, and these were classified into 242 named species and 83 potentially novel species. Of note, 32 of the 89 (36.0%) known species revealed by PacBio sequencing were found to be pathogenic bacteria, and 69 of the 242 (28.5%) known species isolated by culturomics were harmful to people, animals, or plants. Additionally, nearly 40 potential novel species which may be potential bacterial pathogens were identified. IMPORTANCE Bats are one of the most diverse and widely distributed groups of mammals living in close proximity to humans. In recent years, bat-borne viruses and the viral zoonotic diseases associated with bats have been studied in great detail. However, the prevalence and abundance of pathogenic bacteria in bats have been largely ignored. This study used high-throughput sequencing techniques (metataxonomics) in combination with traditional culture methods (culturomics) to analyze the bacterial flora in bat feces from different species of bats in China, revealing that bats are natural hosts of pathogenic bacteria and carry many unknown bacteria. The results of this study can be used as guidance for future investigations of bacterial pathogens in bats.

Attentional bias toward health-threat in health anxiety: a systematic review and three-level meta-analysis.

Attentional bias toward health-threat may theoretically contribute to the development and maintenance of health anxiety, but the empirical findings have been controversial. This study aimed to synthesize and explore the heterogeneity in a health-threat related attentional bias of health-anxious individuals, and to determine the theoretical model that better represents the pattern of attentional bias in health anxiety. Four databases (Web of Science, PubMed, PsycINFO, and Scopus) were searched for relevant studies, with 17 articles (N = 1546) included for a qualitative review and 16 articles (18 studies) for a three-level meta-analysis (N = 1490). The meta-analytic results indicated that the health anxiety group, compared to the control group, showed significantly greater attentional bias toward health-threat (g = 0.256). Further analyses revealed that attentional bias type, paradigm, and stimuli type were significant moderators. Additionally, compared to the controls, health-anxious individuals displayed significantly greater attention maintenance (g = 0.327) but nonsignificant attention vigilance to health-threat (g = -0.116). Our results provide evidence for the attention maintenance model in health-anxious individuals. The implications for further research and treatment of elevated health anxiety in the context of coronavirus disease-2019 (COVID-19) were also discussed.

Effects of Environmental Features in Small Public Urban Green Spaces on Older Adults’ Mental Restoration: Evidence from Tokyo

Exposure to small public urban green spaces (SPUGS) has been demonstrated to have mental benefits for older adults. However, studies on identifying the objective environmental features of SPUGS and their effects on mental restoration for older adults remain limited. This study employed a multilevel regression model to investigate the restorative and vitalizing effects of the environmental features of 11 SPUGS in Tokyo. Onsite measurements were conducted in Kita-Ku, and 202 older adults were surveyed. The results showed that: (1) The fitting curve of the green view index and Restoration Outcome Scale (ROS) score showed an inverted U shape—both green view index and boundary enclosure had a strong impact on the mental restoration of older adults;(2) The colorfulness index showed the strongest relationship with the vitalizing effect. (3) The sky view factor and number of seats only influenced the ROS score, while the results of revitalization suggest that large areas of water should be avoided. (4) Physiological Equivalent Temperature (PET) was also confirmed to have negative effects on the mental restoration of older adults in autumn. These empirical findings can be used as a resource to promote the mental health of older adults in the design of SPUGS in high-density Asian countries.

Preliminary Study of the Protectiveness of Vaccination Against the COVID-19 in the Outbreak of VOC Omicron BA.2 - Jilin City, Jilin Province, China, March 3-April 12, 2022.

What is already known about this topic?: An outbreak of coronavirus disease 2019 (COVID-19) of Omicron BA.2 emerged in Jilin City since March 3, 2022, which involved in 27,036 cases by April 12. The vaccination program with inactivated COVID-19 vaccines has been implemented since the beginning of 2021. What is added by this report?: The incidences of moderate, severe, and critical cases in the whole population of the group of 0+1 dose were 1.82-, 9.49-, and 3.85-fold higher than those in the group of 2 doses, and 5.03-, 44.47-, and ∞-fold higher than those received 3 doses vaccination. For the population ≥60 years, the incidences of moderate, severe, and critical cases in the group of 0+1 dose were 29.92, 9.62, and 4.27 per 100,000, showing 4.13-, 43.72-, and 4.85-fold higher than 2 doses, as well as 13.28-, 22.37-, and ∞-fold higher than 3 doses. What are the implications for public health practice?: The incidences of each type of COVID-19 in the population who were fully vaccinated or booster vaccinated in Jilin City were significantly lower than those who were unvaccinated and/or partially vaccinated. Booster vaccination with homologous inactivated vaccines induces stronger protectiveness for COVID-19 caused by variant of concern (VOC) Omicron.

How Do Chinese People View Cyberbullying? A Text Analysis Based on Social Media.

The rise of cyberbullying has been of great concern for the general public. This study aims to explore public attitudes towards cyberbullying on Chinese social media. Cognition and emotion are important components of attitude, and this study innovatively used text analysis to extract the cognition and emotion of the posts. We used a web crawler to collect 53,526 posts related to cyberbullying in Chinese on Sina Weibo in a month, where emotions were detected using the software "Text Mind", a Chinese linguistic psychological text analysis system, and the content analysis was performed using the Latent Dirichlet Allocation topic model. Sentiment analysis showed the frequency of negative emotion words was the highest in the posts; the frequency of anger, anxiety, and sadness words decreased in turn. The topic model analysis identified three common topics about cyberbullying: critiques on cyberbullying and support for its victims, rational expressions of anger and celebrity worship, and calls for further control. In summary, this study quantitatively reveals the negative attitudes of the Chinese public toward cyberbullying and conveys specific public concerns via three common topics. This will help us to better understand the demands of the Chinese public so that targeted support can be proposed to curb cyberbullying.

Exploring potential mechanisms of Suhexiang Pill against COVID-19 based on network pharmacology and molecular docking.

BACKGROUND: The traditional Chinese medicine prescription Suhexiang Pill (SHXP), a classic prescription for the treatment of plague, has been recommended in the 2019 Guideline for coronavirus disease 2019 (COVID-19) diagnosis and treatment of a severe type of COVID-19. However, the bioactive compounds and underlying mechanisms of SHXP for COVID-19 prevention and treatment have not yet been elucidated. This study investigates the mechanisms of SHXP in the treatment of COVID-19 based on network pharmacology and molecular docking. METHODS: First, the bioactive ingredients and corresponding target genes of the SHXP were screened from the traditional Chinese medicine systems pharmacology database and analysis platform database. Then, we compiled COVID-19 disease targets from the GeneCards gene database and literature search. Subsequently, we constructed the core compound-target network, the protein-protein interaction network of the intersection of compound targets and disease targets, the drug-core compound-hub gene-pathway network, module analysis, and hub gene search by the Cytoscape software. The Metascape database and R language software were applied to analyze gene ontology biological processes and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment. Finally, AutoDock software was used for molecular docking of hub genes and core compounds. RESULTS: A total of 326 compounds, 2450 target genes of SHXP, and 251 genes related to COVID-19 were collected, among which there were 6 hub genes of SHXP associated with the treatment of COVID-19, namely interleukin 6, interleukin 10, vascular endothelial growth factor A, signal transducer and activator of transcription 3 (STAT3), tumor necrosis factor (TNF), and epidermal growth factor. Functional enrichment analysis suggested that the effect of SHXP against COVID-19 is mediated by synergistic regulation of several biological signaling pathways, including Janus kinase/ STAT3, phosphatidylinositol 3-kinase (PI3K)-protein kinase B (Akt), T cell receptor, TNF, Nuclear factor kappa-B, Toll-like receptor, interleukin 17, Chemokine, and hypoxia-inducible factor 1 signaling pathways. SHXP may play a vital role in the treatment of COVID-19 by suppressing the inflammatory storm, regulating immune function, and resisting viral invasion. Furthermore, the molecular docking results showed an excellent binding affinity between the core compounds and the hub genes. CONCLUSION: This study preliminarily predicted the potential therapeutic targets, signaling pathways, and molecular mechanisms of SHXP in the treatment of severe COVID-19, which include the moderate immune system, relieves the "cytokine storm," and anti-viral entry into cells.

Incidence, Etiology, and Environmental Risk Factors of Community-Acquired Pneumonia Requiring Hospitalization in China: A 3-Year, Prospective, Age-Stratified, Multicenter Case-Control Study.

Background: Community-acquired pneumonia (CAP) is a leading infectious cause of hospitalization and death worldwide. Knowledge about the incidence and etiology of CAP in China is fragmented. Methods: A multicenter study performed at 4 hospitals in 4 regions in China and clinical samples from CAP patients were collected and used for pathogen identification from July 2016 to June 2019. Results: A total of 1674 patients were enrolled and the average annual incidence of hospitalized CAP was 18.7 (95% confidence interval, 18.5-19.0) cases per 10000 people. The most common viral and bacterial agents found in patients were respiratory syncytial virus (19.2%) and Streptococcus pneumoniae (9.3%). The coinfections percentage was 13.8%. Pathogen distribution displayed variations within age groups as well as seasonal and regional differences. The severe acute respiratory syndrome coronavirus 2 was not detected. Respiratory virus detection was significantly positively correlated with air pollutants (including particulate matter ≤2.5 µm, particulate matter ≤10 µm, nitrogen dioxide, and sulfur dioxide) and significantly negatively correlated with ambient temperature and ozone content; bacteria detection was opposite. Conclusions: The hospitalized CAP incidence in China was higher than previously known. CAP etiology showed that differences in age, seasons, regions, and respiratory viruses were detected at a higher rate than bacterial infection overall. Air pollutants and temperature have an influence on the detection of pathogens.

Features Constituting Actionable COVID-19 Dashboards: Descriptive Assessment and Expert Appraisal of 158 Public Web-Based COVID-19 Dashboards.

BACKGROUND: Since the outbreak of COVID-19, the development of dashboards as dynamic, visual tools for communicating COVID-19 data has surged worldwide. Dashboards can inform decision-making and support behavior change. To do so, they must be actionable. The features that constitute an actionable dashboard in the context of the COVID-19 pandemic have not been rigorously assessed. OBJECTIVE: The aim of this study is to explore the characteristics of public web-based COVID-19 dashboards by assessing their purpose and users ("why"), content and data ("what"), and analyses and displays ("how" they communicate COVID-19 data), and ultimately to appraise the common features of highly actionable dashboards. METHODS: We conducted a descriptive assessment and scoring using nominal group technique with an international panel of experts (n=17) on a global sample of COVID-19 dashboards in July 2020. The sequence of steps included multimethod sampling of dashboards; development and piloting of an assessment tool; data extraction and an initial round of actionability scoring; a workshop based on a preliminary analysis of the results; and reconsideration of actionability scores followed by joint determination of common features of highly actionable dashboards. We used descriptive statistics and thematic analysis to explore the findings by research question. RESULTS: A total of 158 dashboards from 53 countries were assessed. Dashboards were predominately developed by government authorities (100/158, 63.0%) and were national (93/158, 58.9%) in scope. We found that only 20 of the 158 dashboards (12.7%) stated both their primary purpose and intended audience. Nearly all dashboards reported epidemiological indicators (155/158, 98.1%), followed by health system management indicators (85/158, 53.8%), whereas indicators on social and economic impact and behavioral insights were the least reported (7/158, 4.4% and 2/158, 1.3%, respectively). Approximately a quarter of the dashboards (39/158, 24.7%) did not report their data sources. The dashboards predominately reported time trends and disaggregated data by two geographic levels and by age and sex. The dashboards used an average of 2.2 types of displays (SD 0.86); these were mostly graphs and maps, followed by tables. To support data interpretation, color-coding was common (93/158, 89.4%), although only one-fifth of the dashboards (31/158, 19.6%) included text explaining the quality and meaning of the data. In total, 20/158 dashboards (12.7%) were appraised as highly actionable, and seven common features were identified between them. Actionable COVID-19 dashboards (1) know their audience and information needs; (2) manage the type, volume, and flow of displayed information; (3) report data sources and methods clearly; (4) link time trends to policy decisions; (5) provide data that are "close to home"; (6) break down the population into relevant subgroups; and (7) use storytelling and visual cues. CONCLUSIONS: COVID-19 dashboards are diverse in the why, what, and how by which they communicate insights on the pandemic and support data-driven decision-making. To leverage their full potential, dashboard developers should consider adopting the seven actionability features identified.

A Novel Prediction Method Based on Artificial Intelligence and Internet of Things for Detecting Coronavirus Disease (COVID-19)

Novel coronavirus spreads fast and has a huge impact on the whole world. In light of the spread of novel coronaviruses, we develop one big data prediction model of novel coronavirus epidemic in the context of intelligent medical treatment, taking into account all factors of infection and death and implementing emerging technologies, such as the Internet of Things (IoT) and machine learning. Based on the different application characteristics of various machine learning algorithms in the medical field, we propose one artificial intelligence prediction model based on random forest. Considering the loose coupling between the data preparation stage and the model training stage, such as data collection and data cleaning in the early stage, we adopt the IoT platform technology to integrate the data collection, data cleaning, machine learning training model, and front- and back-end frameworks to ensure the tight coupling of each module. To validate the proposed prediction model, we perform the evaluation work. In addition, the performance of the prediction model is analyzed to ensure the information accuracy of the prediction platform.

Discovery and Evolution of a Divergent Coronavirus in the Plateau Pika From China That Extends the Host Range of Alphacoronaviruses.

Although plateau pikas are the keystone species in the plateau ecosystem of the Qinghai Province of China, little is known about their role in the evolution and transmission of viral pathogens, especially coronaviruses. Here, we describe the characterization and evolution of a novel alphacoronavirus, termed plateau pika coronavirus (PPCoV) P83, which has a prevalence of 4.5% in plateau pika fecal samples. In addition to classical gene order, the complete viral genome contains a unique nonstructural protein (NS2), several variable transcription regulatory sequences and a highly divergent spike protein. Phylogenetic analysis indicates that the newly discovered PPCoV falls into the genus Alphacoronavirus and is most closely related to rodent alphacoronaviruses. The co-speciation analysis shows that the phylogenetic trees of the alphacoronaviruses and their hosts are not always matched, suggesting inter-species transmission is common in alphacoronaviruses. And, PPCoV origin was estimated by molecular clock based on membrane and RNA-dependent RNA polymerase encoding genes, respectively, which revealed an apparent discrepancy with that of co-speciation analysis. PPCoV was detected mainly in intestinal samples, indicating a potential enteric tropism for the virus. Overall, this study extends the host range of alphacoronaviruses to a new order (Lagomorpha), indicating that plateau pikas may be the natural reservoir of PPCoV and play an important and long-term role in alphacoronavirus evolution.

Large-scale integration of meta-QTL and genome-wide association study discovers the genomic regions and candidate genes for yield and yield-related traits in bread wheat.

KEY MESSAGE: Based on the large-scale integration of meta-QTL and Genome-Wide  Association Study, 76 high-confidence MQTL regions and 237 candidate genes that affected wheat yield and yield-related traits were discovered. Improving yield and yield-related traits are key goals in wheat breeding program. The integration of accumulated wheat genetic resources provides an opportunity to uncover important genomic regions and candidate genes that affect wheat yield. Here, a comprehensive meta-QTL analysis was conducted on 2230 QTL of yield-related traits obtained from 119 QTL studies. These QTL were refined into 145 meta-QTL (MQTL), and 89 MQTL were verified by GWAS with different natural populations. The average confidence interval (CI) of these MQTL was 2.92 times less than that of the initial QTL. Furthermore, 76 core MQTL regions with a physical distance less than 25 Mb were detected. Based on the homology analysis and expression patterns, 237 candidate genes in the MQTL involved in photoperiod response, grain development, multiple plant growth regulator pathways, carbon and nitrogen metabolism and spike and flower organ development were determined. A novel candidate gene TaKAO-4A was confirmed to be significantly associated with grain size, and a CAPS marker was developed based on its dominant haplotype. In summary, this study clarified a method based on the integration of meta-QTL, GWAS and homology comparison to reveal the genomic regions and candidate genes that affect important yield-related traits in wheat. This work will help to lay a foundation for the identification, transfer and aggregation of these important QTL or candidate genes in wheat high-yield breeding.

The SARS-CoV-2 nucleocapsid phosphoprotein forms mutually exclusive condensates with RNA and the membrane-associated M protein.

The multifunctional nucleocapsid (N) protein in SARS-CoV-2 binds the ~30 kb viral RNA genome to aid its packaging into the 80-90 nm membrane-enveloped virion. The N protein is composed of N-terminal RNA-binding and C-terminal dimerization domains that are flanked by three intrinsically disordered regions. Here we demonstrate that the N protein's central disordered domain drives phase separation with RNA, and that phosphorylation of an adjacent serine/arginine rich region modulates the physical properties of the resulting condensates. In cells, N forms condensates that recruit the stress granule protein G3BP1, highlighting a potential role for N in G3BP1 sequestration and stress granule inhibition. The SARS-CoV-2 membrane (M) protein independently induces N protein phase separation, and three-component mixtures of N + M + RNA form condensates with mutually exclusive compartments containing N + M or N + RNA, including annular structures in which the M protein coats the outside of an N + RNA condensate. These findings support a model in which phase separation of the SARS-CoV-2 N protein contributes both to suppression of the G3BP1-dependent host immune response and to packaging genomic RNA during virion assembly.

Beta- and Novel Delta-Coronaviruses Are Identified from Wild Animals in the Qinghai-Tibetan Plateau, China.

Outbreaks of severe virus infections with the potential to cause global pandemics are increasingly concerning. One type of those commonly emerging and re-emerging pathogens are coronaviruses (SARS-CoV, MERS-CoV and SARS-CoV-2). Wild animals are hosts of different coronaviruses with the potential risk of cross-species transmission. However, little is known about the reservoir and host of coronaviruses in wild animals in Qinghai Province, where has the greatest biodiversity among the world's high-altitude regions. Here, from the next-generation sequencing data, we obtained a known beta-coronavirus (beta-CoV) genome and a novel delta-coronavirus (delta-CoV) genome from faecal samples of 29 marmots, 50 rats and 25 birds in Yushu Tibetan Autonomous Prefecture, Qinghai Province, China in July 2019. According to the phylogenetic analysis, the beta-CoV shared high nucleotide identity with Coronavirus HKU24. Although the novel delta-CoV (MtCoV) was closely related to Sparrow deltacoronavirus ISU42824, the protein spike of the novel delta-CoV showed highest amino acid identity to Sparrow coronavirus HKU17 (73.1%). Interestingly, our results identified a novel host (Montifringilla taczanowskii) for the novel delta-CoV and the potential cross-species transmission. The most recent common ancestor (tMRCA) of MtCoVs along with other closest members of the species of Coronavirus HKU15 was estimated to be 289 years ago. Thus, this study increases our understanding of the genetic diversity of beta-CoVs and delta-CoVs, and also provides a new perspective of the coronavirus hosts.

Risk Spillover Network Structure Learning for Correlated Financial Assets: A Directed Acyclic Graph Approach

Using cross-asset return data in global financial markets, we propose a novel empirical framework to identify the causal structure of the asset risk spillover network. The joint return distribution of the global financial system can be characterized using a directed acyclic graph approach. However, since assets tend to be highly correlated during market turbulence, when adopting a nodewise penalized regression approach for neighborhood estimation, parameter estimates will receive large standard errors, and edges cannot be reliably estimated. In this work, we propose a two-stage approach for directed acyclic graph skeleton estimation for highly correlated variables. In the first stage, a variable screening ensemble is incorporated into the sparse partial least squares regression method to both reduce the size of the active variables set and impose an adaptive penalization on the weight vectors. In the second stage, a modified PC algorithm based on Gram-Schmidt orthogonalization is applied to remove the false positive edges. Simulation studies are conducted to demonstrate the effectiveness of the proposed method. Finally, we apply our method to analyze the asset risk spillover channels for international financial assets during the COVID-19 pandemic.