TL;DR:
- Scripps Research scientists have developed a powerful AI-based machine-learning system to track the evolution of epidemic viruses and predict emerging viral variants.
- The system successfully predicted the emergence of new SARS-CoV-2 “variants of concern” (VOCs) ahead of their official designations by the WHO.
- The software, using Gaussian process-based spatial covariance, analyzed genetic sequences, variant frequencies, and COVID-19 mortality rates to identify key gene variants driving changes in viral spread and mortality.
- By tracking thousands of undesignated variants, the system serves as an early warning “anomaly detector” for significant shifts in pandemic trajectories.
- The AI system has potential applications beyond tracking pandemics, aiding the development of treatments and vaccines by understanding virus biology.
Main AI News:
In a groundbreaking development, scientists at Scripps Research have harnessed the power of machine-learning, a form of Artificial Intelligence (AI) application, to revolutionize our understanding of viral pandemics. Their cutting-edge system tracks the intricate evolution of epidemic viruses and forecasts the emergence of critical new viral variants. This groundbreaking research, unveiled in a recent paper published in Cell Patterns on July 21, 2023, showcases the system’s remarkable potential by analyzing recorded SARS-CoV-2 variants and COVID-19 mortality rates. The findings suggest that this novel AI-based approach could have foreseen the emergence of new SARS-CoV-2 “variants of concern” (VOCs) even before the World Health Organization (WHO) officially designated them as such. With the potential to track future viral pandemics in real-time, this system holds tremendous promise for global public health.
Dr. William Balch, the senior author of the study and a renowned professor in the Department of Molecular Medicine at Scripps Research, shares his enthusiasm about this groundbreaking approach, “There are rules of pandemic virus evolution that we have not understood but can be discovered, and used in an actionable sense by private and public health organizations, through this unprecedented machine-learning approach.“
Leading the study were co-first authors Dr. Salvatore Loguercio, a former staff scientist at the Balch lab and currently a staff scientist at the Scripps Research Translational Institute, and Dr. Ben Calverley, a postdoctoral research associate in the Balch lab. The Balch lab, well-versed in developing computational methods, particularly AI-based techniques, to understand how genetic variations impact diseases’ symptoms and transmission, applied their expertise to the COVID-19 pandemic.
Leveraging a cutting-edge machine-learning software utilizing a technique known as Gaussian process-based spatial covariance, the researchers combined three essential datasets: genetic sequences of SARS-CoV-2 variants from infected individuals globally, the frequency of these variants, and the global COVID-19 mortality rate. Despite sourcing data from publicly available repositories, the software’s versatility allows it to integrate with any genetic mapping resource.
This powerful software facilitated the tracking of genetic changes in SARS-CoV-2 variants worldwide. These changes exhibited trends such as increased spread rates and decreased mortality rates, signifying the virus’s adaptations to various factors like lockdowns, mask-wearing, vaccines, growing natural immunity, and intervariant competition.
Dr. Balch notes, “We could see key gene variants appearing and becoming more prevalent, as the mortality rate also changed, and all this was happening weeks before the VOCs containing these variants were officially designated by the WHO.“
Undoubtedly, this innovative system proves to be an early warning “anomaly detector” for gene variants significantly affecting viral spread and mortality rates. The researchers emphasize the importance of not solely focusing on prominent variants but also considering the vast number of undesignated variants, aptly dubbed the “variant dark matter.”
Remarkably, the potential applications of this system extend beyond tracking pandemics. It could also provide a real-time understanding of virus biology, bolstering the development of treatments and vaccines. Dr. Balch and his team are currently utilizing their AI system to unveil critical insights into how different SARS-CoV-2 proteins interacted during the pandemic’s evolution.
Co-author Dr. Ben Calverley enthusiastically points out, “This system and its underlying technical methods have many possible future applications.”
“Understanding the Host-Pathogen Evolutionary Balance through Gaussian Process Modelling of SARS-CoV-2,” co-authored by Salvatore Loguercio, Ben Calverley, Chao Wang, Daniel Shak, Pei Zhao, Shuhong Sun, Scott Budinger, and William Balch, marks a pivotal milestone in the realm of AI-driven research, paving the way for a more proactive and informed approach to combating viral pandemics on a global scale.
Conclusion:
The breakthrough AI-based tracking and early-warning system developed by Scripps Research represents a revolutionary advancement in the fight against viral pandemics. By predicting emerging viral variants and understanding virus evolution, this technology can significantly impact the global public health landscape. Businesses in the healthcare and pharmaceutical sectors can leverage this cutting-edge approach to develop proactive strategies, treatments, and vaccines, thereby gaining a competitive edge in addressing future pandemics effectively. Moreover, the application of AI in virus research showcases the immense potential of advanced analytics in driving innovation and disruption in various markets, presenting new opportunities for investors and businesses to invest in transformative technologies.
