Using machine learning to detect coronaviruses potentially infectious to humans.

Establishing the host range for novel viruses remains a challenge. Here, we address the challenge of identifying non-human animal coronaviruses that may infect humans by creating an artificial neural network model that learns from spike protein sequences of alpha and beta coronaviruses and their binding annotation to their host receptor. The proposed method produces a human-Binding Potential (h-BiP) score that distinguishes, with high accuracy, the binding potential among coronaviruses. Three viruses, previously unknown to bind human receptors, were identified: Bat coronavirus BtCoV/133/2005 and Pipistrellus abramus bat coronavirus HKU5-related (both MERS related viruses), and Rhinolophus affinis coronavirus isolate LYRa3 (a SARS related virus). We further analyze the binding properties of BtCoV/133/2005 and LYRa3 using molecular dynamics. To test whether this model can be used for surveillance of novel coronaviruses, we re-trained the model on a set that excludes SARS-CoV-2 and all viral sequences released after the SARS-CoV-2 was published. The results predict the binding of SARS-CoV-2 with a human receptor, indicating that machine learning methods are an excellent tool for the prediction of host expansion events.

The potential of genomics for infectious disease forecasting.

Genomic technologies have led to tremendous gains in understanding how pathogens function, evolve and interact. Pathogen diversity is now measurable at high precision and resolution, in part because over the past decade, sequencing technologies have increased in speed and capacity, at decreased cost. Alongside this, the use of models that can forecast emergence and size of infectious disease outbreaks has risen, highlighted by the coronavirus disease 2019 pandemic but also due to modelling advances that allow for rapid estimates in emerging outbreaks to inform monitoring, coordination and resource deployment. However, genomics studies have remained largely retrospective. While they contain high-resolution views of pathogen diversification and evolution in the context of selection, they are often not aligned with designing interventions. This is a missed opportunity because pathogen diversification is at the core of the most pressing infectious public health challenges, and interventions need to take the mechanisms of virulence and understanding of pathogen diversification into account. In this Perspective, we assess these converging fields, discuss current challenges facing both surveillance specialists and modellers who want to harness genomic data, and propose next steps for integrating longitudinally sampled genomic data with statistical learning and interpretable modelling to make reliable predictions into the future.

Analyzing COVID-19 Using Multisource Data: An Integrated Approach of Visualization, Spatial Regression, and Machine Learning.

Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2, was first identified in Wuhan, China, in December 2019. As the number of COVID-19 infections and deaths worldwide continues to increase rapidly, the prevention and control of COVID-19 remains urgent. This article aims to analyze COVID-19 from a geographical perspective, and this information can provide useful insights for rapid visualization of spatial-temporal epidemic information and identification of the factors important to the spread of COVID-19. A new type of vitalization method, called the point grid map, is integrated with calendar-based visualization to show the spatial-temporal variations in COVID-19. The combination of mixed geographically weighted regression (mixed GWR) and extreme gradient boosting (XGBoost) is used to identify the potential factors and the corresponding importance. The visualization results clearly reflect the spatial-temporal patterns of COVID-19. The quantified results reveal that the impact of population outflow from Wuhan is the most important factor and indicate statistically significant spatial heterogeneity. Our results provide insights into how multisource big geodata can be employed within the framework of integrating visualization and analytical methods to characterize COVID-19 trends. In addition, this work can help understand the influential factors for controlling and preventing epidemics, which is important for policy design and effective decision-making for controlling COVID-19. The results reveal that one of the most effective ways to control COVID-19 include controlling the source of infection, cutting off the transmission route, and protecting vulnerable groups.

Modelling the impact of household size distribution on the transmission dynamics of COVID-19.

Under the implementation of non-pharmaceutical interventions such as social distancing and lockdowns, household transmission has been shown to be significant for COVID-19, posing challenges for reducing incidence in settings where people are asked to self-isolate at home and to spend increasing amounts of time at home due to distancing measures. Accordingly, characteristics of households in a region have been shown to relate to transmission heterogeneity of the virus. We introduce a discrete-time stochastic epidemiological model to examine the impact of the household size distribution in a region on the transmission dynamics. We choose parameters to reflect incidence in two health regions of the Greater Vancouver area in British Columbia and simulate the impact of distancing measures on transmission, with household size distribution the only different parameter between simulations for the two regions. Our result suggests that the dissimilarity in household size distribution alone can cause significant differences in incidence of the two regions, and the distributions drive distinct dynamics that match reported cases. Furthermore, our model suggests that offering individuals a place to isolate outside their household can speed the decline in cases, and does so more effectively where there are more larger households.