A year after the start of the COVID-19 pandemic, mass vaccinations have begun to raise the tantalizing prospect of herd immunity that eventually curbs or stops the spread of SARS-CoV-2. But what if herd immunity is never fully achieved – or if the mutant virus gives rise to hyper-virulent variants that diminish the benefits of vaccination?
These questions highlight the need for effective treatments for people who continue to fall ill with the coronavirus. Although a few existing drugs have advantages, there is an urgent need to find new treatments.
Led by Tudor Oprea, MD, PhD, University of New Mexico, scientists have created a unique tool to help drug researchers quickly identify molecules capable of disarming the virus before it invades cells humans or turn it off in the early stages of infection.
In an article published this week in Artificial intelligence of nature, the researchers presented REDIAL-2020, an open-source online suite of computational models that will help scientists quickly screen small molecules for their potential COVID-fighting properties.
“To a certain extent, this replaces (lab) experiments,” says Oprea, head of the translational informatics division of the UNM medical school. It narrows down the field people have to focus on. That’s why we put it online for anyone to use. “
The Oprea team at UNM and another group from the University of Texas at El Paso led by Suman Sirimulla, PhD, began work on the REDIAL-2020 tool last spring after scientists at the National Center for Advancing Translational Sciences (NCATS) have published their own data. COVID Drug Reuse Studies.
“Realizing this, I was like, ‘Wait a minute, there’s enough data in here for us to be able to build strong machine learning models,’” says Oprea. NCATS lab test results measured the ability of each molecule to inhibit viral entry, infectivity, and reproduction, like the cytopathic effect – the ability to protect a cell from being killed by the virus.
Biomedicine researchers often tend to focus on the positive results of their studies, but in this case, scientists at NCATS also reported which molecules had no anti-virus effects. Including negative data actually improves the accuracy of machine learning, Oprea says.
“The idea was for us to identify molecules that match the perfect profile,” he says. “You want to find molecules that do all of these things and don’t do the things we don’t want them to do.”
The coronavirus is a cunning opponent, says Oprea. “I don’t think there is one drug that works well for a T.” Instead, researchers will likely design a multi-drug cocktail that attacks the virus on multiple fronts. “It goes back to the one-two punch,” he says.
REDIAL-2020 is based on machine learning algorithms capable of rapidly processing huge amounts of data and uncovering hidden patterns that might not be noticeable by a human researcher. The Oprea team validated the machine learning predictions based on NCATS data by comparing them to the known effects of approved drugs in the UNM DrugCentral database.
In principle, this computational workflow is flexible and could be trained to assess compounds against other pathogens, as well as to assess chemicals that have not yet been approved for human use, says Oprea.
“Our main focus is still drug reuse, but we’re actually focusing on any small molecule,” he says. “It doesn’t have to be an approved drug. Anyone who tests their molecule can find something important.”
Source of the story:
Material provided by University of New Mexico Health Sciences Center. Original written by Michael Haederle. Note: Content can be changed for style and length.