According to new research, an artificial intelligence (AI) algorithm can identify which strains of the SARS-CoV-2 virus will likely cause subsequent waves of infection. Further, as per the researchers, the model can discover approximately 73% of variants in each country that will produce at least 1,000 instances per 10 lakh people in three months after a one-week observation period, and more than 80% after two weeks.

The researchers from the Massachusetts Institute of Technology and The Hebrew University-Hadassah Medical School in Israel examined 9 SARS-CoV-2 million genomic sequences collected by the Global Initiative for Sharing Avian Influenza Data (GISAID) from 30 nations. This information was merged with information on vaccination rates, infection rates, and other factors. The goal of the project "promotes the rapid sharing of data from priority pathogens including influenza, hCoV-19, respiratory syncytial virus (RSV), hMpxV as well as arboviruses including chikungunya, dengue and zika," according to the agency's website.

The patterns discovered throughout the investigation were utilized to create a risk assessment model based on machine-learning, an AI algorithm that can learn from prior data and generate predictions. Their findings have been published in the journal PNAS Nexus. The researchers discovered that among the characteristics impacting a variant's infectiousness, the early trajectory of the infections it produced, its spike mutations, and how dissimilar its mutations were from those of the most dominant variant during the observation period were the strongest predictors.

"These results support the hypothesis that the infectious new variants are those that acquire enough mutations which either can lead to reinfections or enable targeting new subgroups of the population that were naturally immune to previous variants," the researchers noted at the time of their study.

They claim that current models for predicting viral transmission dynamics and trends do not account for variant-specific propagation. According to the authors of the study, "this study leverages variant-specific genetic data together with epidemiological information to provide improved early signals and predict the future spread of newly detected variants."

The unique modeling approach might be expanded to other respiratory viruses such as Influenza, Avian Flu viruses, or other Coronaviruses, as well as predict the future course of other infectious disorders. The researchers say that future research might look at how genetic and biological understanding of a variant's infectiousness and distribution can be converted into predictive characteristics that can be evaluated using accessible data.