Predicting unseen antibodies’ neutralizability via adaptive graph neural networks

The effects of novel antibodies are hard to predict owing to the complex interactions between antibodies and antigens. Zhang and colleagues use a graph-based method to learn a dynamic representation that allows for predictions of neutralization activity and demonstrate the method by recommending probable antibodies for human immunodeficiency virus, severe acute respiratory syndrome coronavirus 2, influenza and dengue. Most natural and synthetic antibodies are 'unseen'. That is, the demonstration of their neutralization effects with any antigen requires laborious and costly wet-lab experiments. The existing methods that learn antibody representations from known antibody-antigen interactions are unsuitable for unseen antibodies owing to the absence of interaction instances. The DeepAAI method proposed herein learns unseen antibody representations by constructing two adaptive relation graphs among antibodies and antigens and applying Laplacian smoothing between unseen and seen antibodies' representations. Rather than using static protein descriptors, DeepAAI learns representations and relation graphs 'dynamically', optimized towards the downstream tasks of neutralization prediction and 50% inhibition concentration estimation. The performance of DeepAAI is demonstrated on human immunodeficiency virus, severe acute respiratory syndrome coronavirus 2, influenza and dengue. Moreover, the relation graphs have rich interpretability. The antibody relation graph implies similarity in antibody neutralization reactions, and the antigen relation graph indicates the relation among a virus's different variants. We accordingly recommend probable broad-spectrum antibodies against new variants of these viruses.

Investigation of the Mechanism of Periploca forrestii against Rheumatoid Arthritis with Network Pharmacology-Based Analysis.

Periploca forrestii Schltr (P. forrestii) is an edible medicinal herb with various health benefits such as treating antirheumatoid arthritis (RA), reducing inflammation, and preventing tumor growth. The active ingredients in P. forrestii responsible for its protective effect against RA, however, remain unknown. In this study, the active ingredient of P. forrestii and its potential mechanism of action against RA were investigated by network pharmacology and enrichment analysis. The methods included predicting target genes of P. forrestii, constructing a protein interaction network, and performing gene-ontology (GO) and Kyoto-encyclopedia of genes and genomes (KEGG) enrichment analysis. We discovered targets of RA through retrieval of OMIM and GeneCards public databases. Cardiac glycosides (CGs) are considered the primarily active ingredients of P. forrestii, and the target genes of GCs were discovered to be overlapped with relevant targets of RA using the Venn diagram. After that, prediction of relevant targets of P. forrestii was accomplished with a network pharmacology-based approach. Through the Venn diagram, we discovered 99 genes shared in the target genes of P. forrestii and RA. Gene enrichment analysis showed that the mechanisms of CGs against RA are associated with 55 signaling pathways, including endocrine resistance, Epstein-Barr virus infection, bladder cancer, prostate cancer, and coronavirus disease (COVID-19) signaling pathways. Coexpression analysis indicated ADSL, ATIC, AR, CCND1, MDM2, and HSP90AA1 as the hub genes between putative targets of P. forrestii-derived CGs and known therapeutic targets of RA. In conclusion, we clarified the mechanism of action of P. forrestii against RA, which would provide a basis for further understanding the clinical application of P. forrestii.

COVID-19 Vaccine intent in appalachian patients with multiple sclerosis.

BACKGROUND: Rural people with Multiple Sclerosis (PwMS) face distinctive challenges in the COVID-19 pandemic. The purpose of this study was to determine the COVID-19 vaccine intent and factors associated with vaccine hesitancy among Appalachian adults with MS. METHOD: We conducted a cross sectional phone and in-person survey of PwMS in a large academic center in West Virginia (WV) from February to May 2021. The study sample consists of 306 adult participants. RESULTS: Among the 306 participants, 104 (33.99%) indicated vaccine hesitancy. Statistically significant factors (p<0.05) associated with vaccine hesitancy compared to those who received or intend to get vaccinated included concerns about vaccine safety, vaccine causing MS relapse, vaccine making MS medication ineffective, vaccine causing other diseases, getting the COVID-19 infection, vaccine fast approval, vaccine ingredients, how well the vaccine works, and its side-effects. Additional factors included prior bad experiences with other vaccines, history of not getting the flu vaccine, and lack of consultation about COVID-19 vaccine with healthcare providers. CONCLUSIONS: Vaccine hesitancy among Appalachian adult PwMS is higher compared to PwMS in the larger United States. Vaccine hesitancy is especially higher among those who are female, younger than 50 years old, and residing in rural areas. Concerns about vaccine safety, perception of infection risks, past vaccine behaviors and consultation with healthcare providers are important factors associated with vaccine intent. Factors influencing vaccine hesitancy in Appalachian PwMS are largely consistent with the general public, however, concerns for interaction between the vaccine and MS are specific to this population and thus could be the focus of further vaccine effort.

Improving influenza surveillance based on multi-granularity deep spatiotemporal neural network.

Influenza is a common respiratory disease that can cause human illness and death. Timely and accurate prediction of disease risk is of great importance for public health management and prevention. The influenza data belong to typical spatiotemporal data in that influenza transmission is influenced by regional and temporal interactions. Many existing methods only use the historical time series information for prediction, which ignores the effect of spatial correlations of neighboring regions and temporal correlations of different time periods. Mining spatiotemporal information for risk prediction is a significant and challenging issue. In this paper, we propose a new end-to-end spatiotemporal deep neural network structure for influenza risk prediction. The proposed model mainly consists of two parts. The first stage is the spatiotemporal feature extraction stage where two-stream convolutional and recurrent neural networks are constructed to extract the different regions and time granularity information. Then, a dynamically parametric-based fusion method is adopted to integrate the two-stream features and making predictions. In our work, we demonstrate that our method, tested on two influenza-like illness (ILI) datasets (US-HHS and SZ-HIC), achieved the best performance across all evaluation metrics. The results imply that our method has outstanding performance for spatiotemporal feature extraction and enables accurate predictions compared to other well-known influenza forecasting models.

A safety consideration of mesenchymal stem cell therapy on COVID-19.

Due to the multi-potential differentiation and immunomodulatory function, mesenchymal stem cells (MSCs) have been widely used in the therapy of chronic and autoimmune diseases. Recently, the novel coronavirus disease 2019 (COVID-19) has grown to be a global public health emergency but no effective drug is available to date. Several studies investigated MSCs therapy for COVID-19 patients. However, it remains unclear whether MSCs could be the host cells of SARS-CoV-2 (severe acute respiratory syndrome coronavirus-2) and whether they might affect the SARS-CoV-2 entry into other cells. Here, we report that human MSCs barely express ACE2 and TMPRSS2, two receptors required for the virus endocytosis, indicating that MSCs are free from SARS-CoV-2 infection. Furthermore, we observed that MSCs were unable to induce the expression of ACE2 and TMPRSS2 in epithelial cells and macrophages. Importantly, under different inflammatory challenge conditions, implanted human MSCs failed to up-regulate the expression of ACE2 and TMPRSS2 in the lung tissues of mice. Intriguingly, we showed that a SARS-CoV-2 pseudovirus failed to infect MSCs and co-cultured MSCs did not increase the risk of SARS-CoV-2 pseudovirus infection in epithelial cells. All these results suggest that human MSCs have no risk of assisting SARS-CoV-2 infection and the use of MSCs as the therapy for COVID-19 patients is feasible and safe.