Biotechnology

Scripps Research scientists develop AI-based tracking and early warning

[ad_1]

LA JOLLA, CA — Scripps Research scientists have developed a machine learning system—a type of artificial intelligence (AI) application—that can track the detailed evolution of an epidemic virus and predict the emergence of viral variants with important new traits.

Credit: Image created using BioRender.com.

LA JOLLA, CA — Scripps Research scientists have developed a machine learning system—a type of artificial intelligence (AI) application—that can track the detailed evolution of an epidemic virus and predict the emergence of viral variants with important new traits.

In the paper at Cell Pattern on July 21, 2023, scientists demonstrated the system using data on recorded variants of SARS-CoV-2 and death rates from COVID-19. They demonstrated that the system could predict the emergence of new SARS-CoV-2 “concern variants” before their official designation by the World Health Organization (WHO). Their findings point to the possibility of using such a system in real-time to track future viral pandemics.

“There are rules of pandemic virus evolution that we don’t yet understand but that can be discovered, and used in an actionable sense by private and public health organizations, through this unprecedented machine learning approach,” said study senior author William Balch, PhD, professor in the Department of Molecular Medicine at Scripps Research.

The first co-author of the study was Salvatore Loguercio, PhD, staff scientist in the Balch lab at the time of the study, and currently a staff scientist at the Scripps Research Translational Institute; and Ben Calverley, PhD, postdoctoral research fellow in the Balch lab.

Balch’s lab specializes in developing computational, often AI-based methods to explain how genetic variations change symptoms and the spread of disease. For this study, they applied their approach to the COVID-19 pandemic. They developed machine learning software, using a strategy called Gaussian process-based spatial covariance, to correlate three data sets covering the course of the pandemic: the genetic sequence of the SARS-CoV-2 variants found in infected people around the world, the frequency of those variants, and the global death rate for COVID-19.

“This computational method uses data from publicly available repositories,” said Loguercio. “But it can be applied to any genetic mapping resource.”

The software allows researchers to track the sequence of genetic changes that appear in variants of SARS-CoV-2 around the world. These changes—usually leading to increased prevalence rates and decreased mortality rates—signify the adaptation of the virus to lockdowns, wearing masks, vaccines, boosting natural immunity in the global population, and relentless competition among the variants of SARS-CoV-2 themselves.

“We could see key gene variants emerge and become more common, as mortality rates also changed, and all this was weeks before the VOC containing these variants was officially designated by WHO,” said Balch.

He and his team demonstrated that they could use this SARS-CoV-2 tracking system as an early warning “anomaly detector” for gene variants associated with significant changes in viral spread and mortality rates.

“One of the big lessons from this work is that it is important to consider not only a few prominent variants, but also tens of thousands of other unspecified variants, which we call ‘dark matter variants,’” Balch said.

A similar system could be used to track the detailed evolution of future viral pandemics in real time, the researchers note. In principle, this would enable scientists to predict changes in the trajectory of a pandemic—for example, a large increase in infection rates—in time to implement appropriate public health precautions.

Balch and his colleagues also envision using their approach to better understand virus biology and thereby improve treatment and vaccine development. Now they are using their AI systems to uncover key details about how the various SARS-CoV-2 proteins work together in the evolution of the pandemic.

“This system and the technical methods that underlie it have many possible future applications,” said Calverley.

“Understanding the Host-Pathogen Evolved Balance through Modeling the Gaussian Process of SARS-CoV-2” co-authored by Salvatore Loguercio, Ben Calverley, Chao Wang, Daniel Shak, Pei Zhao, Shuhong Sun, Scott Budinger, and William Balch.

About Scripps Research

Scripps Research is an independent, not-for-profit biomedical organization that has been ranked one of the most influential in the world for its impact on innovation by the Nature Index. We advance human health through in-depth discoveries that address urgent medical problems around the world. Our drug discovery and development division, Calibr, works hand-in-hand with scientists across disciplines to get new drugs to patients as quickly and efficiently as possible, while the team at Scripps Research Translational Institute leverages genomics, digital medicine, and cutting-edge informatics to understand individual health and deliver more effective healthcare. Scripps Research also trains the next generation of leading scientists in our Skaggs Graduate School, which is consistently named among the top 10 US programs for the chemical and biological sciences. Learn more at www.scripps.edu.


[ad_2]

Source link

Related Articles

Back to top button