A new study co-authored by Duane Moser, a microbiologist at the Desert Research Institute (DRI), introduces an artificial intelligence-driven method for detecting and analyzing emerging virus variants through wastewater surveillance. The research builds on previous work that established wastewater monitoring as a cost-effective way to track pathogens during the COVID-19 pandemic.
Moser collaborated with researchers from the University of Nevada, Las Vegas (UNLV) to collect wastewater samples across rural southern Nevada. The study, published in Nature Communications, analyzed 3,659 wastewater samples collected over two years and compared them to 8,810 SARS-CoV-2 clinical genomes from residents in Nevada. This approach revealed differences in how SARS-CoV-2 variants appeared and spread between urban and rural areas, allowing for faster detection of new variants.
The research expands on a 2024 report by the same team that profiled SARS-CoV-2 variants among unhoused populations using water sampled weekly from underground storm drains for more than a year.
“During the pandemic, urban wastewater surveillance proved highly effective in providing information about the timing and severity of community SARS-CoV-2 infection as progressive variants moved through the world’s urban centers,” Moser said. “In contrast, large portions of the world and U.S. represent blank spots on disease tracking maps. My role in the study was to extend the excellent analytical tools being applied in the Las Vegas metropolitan area to the much larger geographic expanse of rural Clark and Nye Counties. During the pandemic, progressive waves of SARS-CoV-2 variants moved across the land – providing a lesson in disease propagation on the landscape scale. Propagation of the pandemic, from initial appearances in urban centers to outlying areas, was almost like clockwork – typically showing a delay of 7 – 9 days. Most variants persisted in these smaller outlying populations for a week or two, only to be replaced by the next and so on. While most of the major variants first recognized in greater Las Vegas did eventually appear in the rural areas, in several cases, a given variant seemed to skip some of the more remote rural locations or perhaps came and went so quickly that our weekly sampling program missed it.”
“Overall, this study validates a growing consensus concerning the power of wastewater surveillance for disease tracking in society, including from low-density rural populations,” he added. “The new tool extends the power of wastewater surveillance to novel pathogen variants that might otherwise go undetected across large geographic expanses in the U.S. and beyond. While the pattern during the Covid pandemic was that of unidirectional outward expansion from urban-to-rural locations, as humanity continues to encroach upon Earth’s wild places, the direction of transmission could just as easily be in the other direction. Thus, wastewater surveillance of outlying areas could enable early detection of the next pandemic before it is released into one of the world’s great cities.”
According to UNLV neuroscience graduate student Xiaowei Zhuang—lead author—the AI-driven algorithm can scan wastewater samples for viruses such as influenza, RSV, mpox, measles, gonorrhea or Candida auris before they are identified by clinical tests.
“Imagine identifying the next outbreak even before the first patient enters a clinic. This research shows how we can make this possible,” said Edwin Oh, professor at UNLV’s Nevada Institute of Personalized Medicine at UNLV and co-author on this study. “Through use of AI we can determine how a pathogen is evolving without even testing a single human being.”
“The tool could be especially useful for improving disease surveillance in rural communities”, said DRI’s Duane Moser who coordinated sampling across 13 outlying wastewater plants featured in this study. “Large portions of U.S. fall outside jurisdiction public health districts thus can represent blank spots on disease tracking maps. The new tool extends power wastewater surveillance novel pathogen variants which might otherwise go undetected across large geographic expanses U.S internationally”.
The team found their AI simulation needed about eight samples per variant to establish reference signatures reliably and as few as two to five samples per variant for early detection.
Traditional methods relied heavily on clinical data from patients already tested but were reactive—often identifying strains after they had begun circulating widely within communities. In contrast, this new approach does not require prior knowledge or patient data; it proactively detects patterns directly from multiple wastewater samples.
“Wastewater surveillance has enabled more timely and proactive public health responses through monitoring disease emergence and spread at a population level in real time,” says Zhuang. “This new method enhances early outbreak detection to allow for identification novel threats without prior knowledge making this tool even more effective public health surveillance moving forward.”
Since 2021 four Las Vegas institutions—UNLV Southern Nevada Water Authority (SNWA), Southern Nevada Health District and DRI—have operated a public dashboard tracking COVID-19 cases via wastewater data.
The Nature Communications AI study is one among more than thirty collaborative studies involving these organizations along with Cleveland Clinic Lou Ruvo Center for Brain Health; researchers say it is first employing an AI approach enhancing what they call "wastewater intelligence".
“Wastewater surveillance has proven to be an effective tool for filling critical data gaps and understanding public health conditions within a community,” said Daniel Gerrity principal research microbiologist at SNWA.“The ongoing wastewater surveillance effort is great example how collaboration between SNWA UNLV other partners can lead positive impacts local community beyond.”