Published December 23, 2020
Researchers at the Jacobs School of Medicine and Biomedical Sciences have developed a computational approach to make drug discovery faster and less expensive while also being safe and effective.
The approach is implemented in a platform called CANDO (Computational Analysis of Novel Drug Opportunities) and was initially funded by a National Institutes of Health Director’s Pioneer Award to generate therapeutic predictions of existing drugs that could be repurposed to rapidly tackle emerging disease outbreaks.
“While the initial focus of CANDO was for shotgun drug repurposing (finding new uses for existing approved drugs), it is capable of novel drug discovery and design as well,” says Ram Samudrala, PhD, professor of biomedical informatics, chief of the Division of Bioinformatics and senior author on the project paper. “Unlike traditional approaches that focus on a single target/disease of interest, CANDO attempts to understand the entire network of interactions between all drugs, all molecules and all diseases simultaneously. It’s a holistic approach.”
Researchers began applying CANDO in the fight against COVID-19 in February and there have already been positive outcomes.
“We used CANDO to suggest already-approved drugs that may in fact be useful for inhibiting the SARS-CoV-2 virus, and therefore be useful for curbing the COVID-19 pandemic,” says William Thomas Mangione, doctoral candidate in the Department of Biomedical Informatics and lead author on the paper.
“Often when scientists try to find therapies for diseases, they consider only one target (usually a protein molecule) to inhibit or modulate, which typically leads to failures in the clinic. Our software considers all available protein targets simultaneously and prioritizes drugs that have strong computed activities against multiple targets,” Mangione adds. “In this case, we calculated interactions between roughly 13,000 drug or drug-like molecules and the 24 SARS-CoV-2 proteins, then suggested those with promising scores.”
“As of now, 10 out of our top 50 or so predictions — which we did back in February — have been validated by our industry partners or by others and shown to be effective against SARS-CoV-2. That is a really high success rate,” Samudrala says.
That is important with much of the country now experiencing a second wave of the pandemic.
“Until vaccines become widely adopted, this just provides more treatment options for those who don’t want or cannot take vaccines,” Samudrala says.
Samudrala says this approach is applicable to any disease.
“Every time there’s a new disease, the platform can screen its library of existing approved drugs to see if any of them could work for them, and every time there’s a new drug developed by traditional methods — which cost billions of dollars and take decades of effort — the platform can see what other diseases it can be repurposed for,” he says.
Zackary Falls, PhD, who was a postdoctoral fellow at the time of the research and is now research assistant professor of biomedical informatics, worked with Mangione in applying CANDO to COVID-19; together they developed CANDO versions 2 and 3 in the research project.
Virologist Thomas Melendy, PhD, associate professor of microbiology and immunology and biochemistry and director of the Witebsky Center for Microbial Pathogenesis and Immunology, provided expert advice on SARS-CoV-2 and the prodrug effect of remdesivir.
Gaurav Chopra, PhD, worked with Samudrala in developing CANDO version 1 as a mentee and is continuing in the research as assistant professor of analytical and physical chemistry in the Department of Chemistry at Purdue University.