The researchers reported in the journal Cell Reports Medicine that the test was 97 percent accurate for SARS-CoV-2 and 95 percent accurate for the most prevalent variant of the hepatitis C virus found globally.
The simplified test developed by scientists at the University of Florida is currently in a tiny test tube, but it could soon reach a doctor’s office and, one day, be available as a home test that’s as easy as a pregnancy test.
“We are trying to create a home-based test that is as reliable as a lab-based test,” said Piyush Jain, professor of chemical engineering who led the latest research.
“Our aim is to develop a simple test that eliminates the need for expensive equipment and provides results in only 10 to 20 minutes,” he added.
Using AI, Jain’s group developed a system known as a one-pot reaction because the entire test takes place in a small test tube. These tests, based on a technology called RT-LAMP, can amplify small sections of the virus’s genome and produce a visual signal when the virus is detected.
Reading these tests can be as simple as looking for a blue color or using a small device that detects changes in a test tube. To determine the difference between false positives and true positives, Jain said he is combining another technique called CRISPR.
However, the RT-LAMP technique requires a temperature of 150°F, while CRISPR works best at 100°F. This difference makes tests more complex that require two separate reactions – too complex for home use.
To bridge the gap, Jain’s group turned to AI tools to analyze a CRISPR enzyme that thrives at 140 degrees and discovered changes to keep it alive at 150 degrees.
“It’s very challenging for any human to do this kind of analysis on a single enzyme. We don’t have to spend years, we can make these improvements in months,” Jain said.
“With everything working at the same temperature, we can now combine everything into a one-pot reaction we call Splendid.”
The team successfully validated their simplified Splendid test on clinical samples from patients with hepatitis C or COVID-19.
While it doesn’t work well against all the other less prominent versions of the hepatitis C virus, Jain said a direct change to the test should rapidly improve its accuracy.