The amount of SARS-CoV-2 in a wastewater system was linked to the prevalence of COVID-19 in the area it served, according to research from the lab of Fangqiong Ling at Washington University in St. Louis earlier this year.
Precision insights can be found in wastewater
Ling, an assistant professor in the Department of Energy, Environmental and Chemical Engineering at the McKelvey School of Engineering, was able to create a machine learning model that used the variety of microbes found in wastewater to determine how many distinct individuals they represented thanks to a chance encounter with a colleague, as per ScienceDaily.
In the future, this technique might be able to connect other wastewater properties to data at the individual level.
Ling initially believed that machine learning could find a clear connection between the diversity of microbes and the number of people it represented, but simulations using "off-the-shelf" machine learning failed to bear this out.
Then Ling met Likai Chen, an assistant professor of statistics and mathematics in Arts & Sciences, by chance.
They discovered they had a similar interest in handling complex, novel data.
Chen might be able to help out on the project Ling is working on, according to Ling.
The ability to determine how many distinct individuals were included in a sample depends on how closely the sample resembles the mean, or average, as sample size increases.
Individuals are not exactly "average," so if a sample of microbiota appears to be representative of many individuals, it probably represents a large number of individuals.
The farther away from the average, the more likely it is to represent an individual.
Together with Ling, Chen used this fundamental intuition to create a more specialized machine learning algorithm that, when trained on actual microbiota samples from over 1,100 people, could determine how many people were represented in a wastewater sample.
Also Read: Biochar Created from Agricultural Waste Adsorbs Wastewater Contaminants
Can I get COVID-19 from wastewater or sewage?
When combined with other data, wastewater surveillance data are most helpful.
In addition to other data, such as overall virus levels in wastewater, historical wastewater data for that location, geographical context (for instance, whether areas have high tourism or neighboring communities with increasing cases), and clinical cases, wastewater data showing the percent change in virus levels should be used, as per the CDC.
As local prevention tactics change, communities may observe changes in the levels of virus wastewater.
Small changes can be detected by early warning systems, such as wastewater surveillance, as a signal for quick action.
It is crucial to remember that when the amount of virus in wastewater is low, a small increase in the overall virus level can appear much larger when converted to percentages.
Small changes can be detected by early warning systems, such as wastewater surveillance, as a signal for quick action.
It is crucial to remember that when the amount of virus in wastewater is low, a small increase in the overall virus level can appear much larger when converted to percentages.
A change from one unit to two units, for instance, would represent a 100% percent change.
A change from 500,000 to 1 million units also represents a 100% percent change.
Related article: Potato Wastewater Could Feed Bacteria to Help Recycle High-Tech Devices