Duo ‘AIMS’ at AI-powered discovery efforts

GENEVA & SAN FRANCISCO—Artificial intelligence (AI) systems provide an invaluable service, especially to academic and non-profit organizations with limited resources. A researcher can identify a target and, with the help of the computational power of AI, can arrive at a short list of potentially therapeutic proteins. Such is the benefit that Atomwise and their Artificial Intelligence Molecular Screen (AIMS) offer as they pair up with non-profit and academic partners in search of novel targets for drug discovery.

A recent win for the AIMS system arose from a partnership with the Drugs for Neglected Diseases initiative (DNDi), a non-profit research organization looking primarily at neglected diseases in the search for therapeutic interventions. The two organizations joined forces and homed in on three verified but challenging therapeutic protein targets that would inhibit the action of the parasite that causes Chagas disease. For each identified disease protein, Atomwise screened millions of compounds using its AI-powered screening technology to predict those that bind and potentially inhibit protein function. Those predictions identify drug-like compounds that can be further optimized and potential developed in to effective drugs.

“This remains an early-stage drug discovery project—the next steps are full validation of the hits identified by Atomwise, followed by hit to lead, lead optimization and eventual nomination of a preclinical candidate,” according to Dr. Benjamin Perry, senior discovery manager and research scientist at DNDi. “From there, the chosen candidate would progress into the development pipeline alongside other DNDi candidate compounds. As in any drug discovery effort, the timelines for this process are difficult to predict, but are on the order of 15 years from the current state of the project.”

Research biologists focus on the biology of disease, identifying key disease proteins. Atomwise then dives into the chemistry side of drug discovery, accessing a library of more than seven million compounds and applying its AI-powered screening technology to predict those that will interact most effectively with the target protein. The vast capacity of the AIMS system is propelling research into Chagas disease, as well as several other ultra-rare neglected diseases that have proven too expensive to garner broad research.

“For DNDi, [AIMS] represents an exceptional way to identify new starting points for drug discovery projects focused on Chagas disease and leishmaniasis,” says Perry. “We have a long history of screening against the causative parasites in a ‘phenotypic’ manner; however, we are keen to also engage in more target-based approaches to tackle these parasites of interest. Atomwise technology provides an excellent and highly efficient alternative to the classical ways of identifying hits for biological targets of interest, such as [high-throughput sequencing] or fragment screening.”

Chagas disease is a parasitic infection affecting more than 6 million people worldwide and is caused by so-called “kissing bugs.” It disproportionately affects poor populations living in pre-modern housing with limited access to healthcare. The disease often lies dormant and unsymptomatic for up to 20 or more years—but when it emerges, it can be lethal. Current treatments for Chagas disease are suboptimal, with long treatment periods and uncertain efficacy for those in the advanced stage of the disease.

“DNDi’s early success in Chagas disease is exemplary of the goal we sought to accomplish when we established the AIMS program to accelerate life-saving drug discovery,” remarked Dr. Abraham Heifets, CEO at Atomwise. “Our scientists are working creatively with DNDi researchers to apply our technology to targets that some might consider undruggable or too early for rational drug design due to insufficient data. While these targets present the greatest challenge, they also provide the greatest opportunity for transformative treatments for Chagas disease. The research conducted through this collaboration could offer a path to novel medicines.”

The AIMS program supports hundreds of researchers at non-profit institutions and universities worldwide at no cost and with no negotiated royalties. For Atomwise, they share the goal of their not-for-profit partners to locate new possibilities, conduct exhaustive testing and publish groundbreaking results that further research across sectors. Atomwise also benefits from being able to positively demonstrate dozens of successes with the AIMS system, generating the data that increases access to more, and more varied, partners.

“Drug discovery is a team sport,” according to Heifets. “There is a ton of hype surrounding artificial intelligence, a lot of noise and excitement. The critical thing for AI practitioners is proof of success, demonstrating over and over again that AI can put the ecosystem in place to create great outcomes and find the right answers.”