At the Drugs for Neglected Diseases initiative (DNDi), Ben Perry has an enviable position. A medicinal chemist by training, Perry now works with leading pharmaceutical companies and academic labs around the world. He also gets to try the latest technologies in his ongoing quest to find new ways of conducting early-stage drug discovery.
Because the targets are neglected diseases, there’s virtually no money to be made from new treatments. “But a lot of companies want to work with us at least a little bit,” Perry says. After all, new insights discovered through a project about Chagas disease could still inform research into other, more lucrative diseases. Academic labs also actively contribute to DNDi by contributing molecules to ongoing optimization projects.
By working with so many stakeholders, Perry and his team get the benefit of a wide variety of approaches. “My personal belief is that the only way we’re going to get better at doing drug discovery is by trying lots of different things,” he says. “Some of those things are going to work and some aren’t. That’s science. But we won’t know unless we try.”
That try-anything concept opened the doors for discussions with AI pioneers at Atomwise. When the company launched its Artificial Intelligence Molecular Screen (AIMS) Award program in 2017, Perry was sold. “We realized that even though we work in neglected diseases, we shouldn’t be playing second fiddle to the big diseases when it comes to trying things at the cutting edge,” he says. “But we rely on public funds and have to be careful about where we put our money. The AIMS Award is a phenomenal idea that let us try out the Atomwise technology.”
The AIMS Award gives scientists free access to customized virtual screening of millions or even billions of small molecules through the company’s AI-powered AtomNet® technology. Atomwise also sends winners dozens of small molecule compounds for follow-up functional testing.
So far, Perry’s team has submitted three Chagas disease targets for the AIMS Award program: one well-known, one less known but relatively straightforward, and one brand-new target expected to be a challenge for virtual screening. The Atomwise team, with projects led by VP of Computer-Aided Drug Design (CADD) Kong Nguyen and Senior CADD Scientist Mostafa Ahmed, got to work. While one target did not lead to new candidate compounds, the other two yielded promising results.
“We know from screening random collections what the hit rate should be,” Perry says. “The hit rate with Atomwise was 10 times higher.”
Perry and his collaborators are following up on those initial results by testing the candidate compounds against Trypanosoma cruzi, the parasite responsible for Chagas disease. They’ve also been working with Atomwise on a second phase of the AIMS Award, which entails subsequent virtual screening to look for closely related compounds that may have better pharmacologic properties. “It’s progressing well and we’re really interested in seeing what the second set of results will be,” Perry says. “They’re even repeating the target where we found nothing new because the AI will have learned more from its iterations on other targets in the past 24 months. It’s got to be one of the most trained and most used AI tools in virtual screening anywhere.”
To learn more about Perry’s research program on Chagas disease, view Atomwise Senior CADD Scientist Mostafa Ahmed’s poster presentation at the American Chemical Society Spring 2021 National Meeting: Artificial Intelligence Aided Discovery of New Classes of Antiprotozoal Agents for Chagas Disease
To learn more about open innovation at DNDi: Open and collaborative drug discovery