A new study conducted by scientists at Scripps Research in California has revealed that an artificial intelligence (AI) machine-learning system has the potential to track the evolution of future pandemics in real time. The system, which was demonstrated using data on recorded COVID-19 variants and mortality rates, could have predicted the emergence of variants of concern before they were officially designated by the World Health Organization (WHO).
The machine-learning system developed by the researchers uses a strategy called Gaussian process-based spatial covariance to analyze three datasets related to the pandemic. These datasets include the genetic sequences of COVID-19 variants, the frequencies of those variants, and the global mortality rate for COVID-19. By analyzing the data, the system was able to identify key gene variants and their increasing prevalence, as well as changes in mortality rates, weeks before the WHO officially recognized them as variants of concern.
According to Professor William Balch, a senior author of the study, there are still rules of pandemic virus evolution that have not been fully understood. However, he believes that this unprecedented machine-learning approach can help private and public health organizations gain actionable insights for tracking and responding to future viral pandemics.
Salvatore Loguercio, a co-author of the study, explained that the computational method used in this research relied on publicly available repositories but could be applied to any genetic mapping resource. This demonstrates the versatility and potential of the machine-learning system for monitoring various viral outbreaks.
The researchers also highlighted the importance of considering not only a few prominent variants but also the tens of thousands of undesignated variants, which they referred to as the variant dark matter. They believe that focusing on this vast pool of genetic variations will contribute to better understanding the evolution of viruses and the associated changes in their spreads and mortality rates.
In addition to its applications in tracking the evolution of pandemics, the AI machine-learning system could serve as an early warning system or anomaly detector for gene variants that are linked to significant changes in viral spreads and mortality rates. By leveraging this technology, scientists and health organizations may be able to predict and respond to emerging variants more effectively.
The findings of this study carry significant implications for the field of public health. By harnessing the power of AI and machine learning, researchers and professionals can gain critical insights into the evolutionary patterns of viruses, helping them to stay one step ahead in the fight against pandemics.
As the world continues to battle the ongoing COVID-19 pandemic and prepare for potential future outbreaks, the development and implementation of advanced technologies like the AI machine-learning system studied by Scripps Research have the potential to revolutionize our approach to infectious diseases. With the ability to predict and track the emergence of new variants, there is hope for more proactive and targeted interventions in the future.
The Scripps Research study serves as a reminder of the importance of utilizing cutting-edge technologies to improve our understanding of viruses and their evolution. By investing in innovative research and finding new ways to harness the power of AI, we can equip ourselves with the necessary tools to minimize the impact of future pandemics and protect global health.
