Experts in AI and drug discovery to converge at CACHE symposium in Toronto

Left to right: John Moult, Judith Guenther, Okexandr Isayev

Can computational chemistry predict medicine-like molecules for any protein?

Maybe, but despite the ability of machine learning to identify such promising compounds, finding a reliable algorithm to select, design and rank these molecules remains a daunting challenge.

To speed up this process, the inaugural CACHE Challenge Symposium will spotlight the role of open science in drug discovery, on March 6 and 7 in Toronto.

CACHE (Critical Assessment of Computational Hit-Finding Experiments) is a public benchmarking project that compares and improves small-molecule hit-finding algorithms, through cycles of prediction and experimental testing that run every four months.

“It’s gratifying to see strong and continued engagement coming both from academia and industry as new CACHE challenges progress,” said Matthieu Schapira, lead of CACHE and chair of the symposium organizing committee, who is a professor of pharmacology and toxicology at the Temerty Faculty of Medicine and principal investigator at the Structural Genomics Consortium.

The symposium will be hosted by Conscience, a new biotech non-profit focused on AI and collaborative science, and the Structural Genomics Consortium. The event’s main theme will be results from the first CACHE Challenge, announced earlier this year by Conscience and covered in a recent perspective in Nature Reviews Drug Discovery.

In that challenge, supported by the Michael J. Fox Foundation, twenty-three academic and commercial organizations competed to identify small molecules that bind the WDR genetic domain of LRRK2 — the most mutated protein in familial Parkinson’s disease.

The challenge uncovered seven new promising hits that could serve as a starting point for treatment research for the disease.

The symposium will feature presentations from the seven best-performing groups in the first challenge, who will share their computational methods and findings, as well as insights on the value of benchmarking competitions from industry leaders, alongside the experimentalist perspective from the Structural Genomics Consortium.

“We are so thrilled to have an array of speakers who will be sharing their experiences in a way that will educate and inspire,” said Ryan Merkley, CEO of Conscience. “Our greatest hope is to foster an environment where innovation and collaboration can flourish.”

Among the speakers, symposium attendees will hear from:

John Moult, a professor at the University of Maryland, who served as co-founder and chair of Critical Assessment of methods of protein Structure Prediction (CASP), the benchmarking challenge that set the stage for the AlphaFold breakthrough.
Judith Guenther, senior data scientist in computational and molecular design at Bayer.
David Koes, an associate professor at the University of Pittsburgh School of Medicine, who developed gnina, the most widely used deep-learning-based open-source virtual screening tool and had one of the top-performing computational methods in the first CACHE Challenge.
Suzanne Ackloo, a project manager at the Structural Genomics Consortium, who coordinates the experimental arm of CACHE at the University of Toronto.
Olexandr Isayev, associate professor at Carnegie Mellon University and a leading pioneer in the application of deep-learning potentials for molecular and materials systems, who tied for the top-performing computational method in the first CACHE Challenge.

The event will also spotlight emerging researchers, offering them a platform to present their work and engage with established scientists in a vibrant poster session.

The CACHE Challenge Symposium will run March 6 and 7 in Toronto. Registration is required, so secure your place now and learn more about the event's full agenda by visiting: https://conscience.ca/cache-challenge-symposium-march-6-7-2024/