Predicting, and being able to prepare for, SARS-CoV-2 variants that will come in the future would be incredibly useful for public health preparedness. To do that, a team has built a new predictive model to allow scientists to attempt to forecast the appearance of potential new mutations in emerging and future variants SARS-CoV-2. The model, which provided accurate forecasts of mutations in the Delta variant, could give scientists and health agencies the means to both predict and manage the constant emergence of new variants of concern.
The work is published in Science Translational Medicine in the article, “Predicting the mutational drivers of future SARS-CoV-2 variants of concern.”
The rapid evolution of SARS-CoV-2 continues to extend the COVID-19 pandemic and deepen its toll, with new variants causing new waves of hospitalizations and deaths.
Although current vaccines still largely protect against severe disease, scientists fear that future variants with new mutations could be even better at evading vaccine-induced immunity. Vaccine efforts and public health preparedness, therefore, depend on being able to understand and predict the emergence and spread of problematic mutations.
Using predictive epidemiological features and neural network-based approaches, researchers created a predictive model to forecast the mutations that might appear in future variants. The researchers tested the predictive value of features comprising epidemiology, evolution, immunology, and neural network-based protein sequence modeling, and identified primary biological drivers of SARS-CoV-2 intra-pandemic evolution.
The team first trained their model with data from previous waves of the pandemic, including preliminary data on mutations in the Omicron variant. The model was able to predict the spike protein mutations that emerged and spread across different phases of the pandemic, with high accuracy and as many as four months in advance.
They found evidence that ACE2-mediated transmissibility and resistance to population-level host immunity has waxed and waned as a primary driver of SARS-CoV-2 evolution over time. They also identified mutations that will spread, up to four months in advance, across different phases of the pandemic.
They then produced a forecast of mutations in the Delta variant and identified several mutations that may contribute to new variants of interest and variants of concern in the coming months, including a mutation that might weaken the effects of clinical therapeutic antibodies.
“This modeling approach may be applied to any rapidly evolving pathogens with sufficiently dense genomic surveillance data, such as influenza, and unknown future pandemic viruses,” the team concluded.
