Israel's AION Labs announced last week the launch of CombinAble.AI, a startup focused on improving antibody design to speed up the development of better therapeutics.
"The launch of CombinAble heralds a shift in the approach to targeted antibody design by leveraging AI to optimize and expedite the discovery process," said Mati Gill, CEO of AION Labs.
AION Labs is an Israeli venture studio operating in partnership with AstraZeneca, Merck, Pfizer, Teva, the Israel Biotech Fund, Amiti Ventures, and Amazon Web Services (AWS). It is supported by BioMed X and the Israel Innovation Authority. Its mission is to develop and integrate innovative artificial intelligence (AI) technologies to revolutionize drug discovery and improve global health.
CombinAble is tackling the drug discovery challenge with a platform that combines advanced AI technology and computational methods for biomolecule simulations, AION Labs explained. Its platform automates the selection of mutations to improve multiple characteristics of antibody candidates, speeding up the discovery and design of new biotherapeutics.
An important stage in drug development
Antibody optimization is a crucial stage in drug development, where a preliminary functional antibody is refined into a potential clinical drug by improving its safety, efficacy, and manufacturability.
Traditionally, this process involves time-consuming trial-and-error methods in the lab, which can take years and incur high costs without guaranteeing success. According to a 2022 report published in Acta Pharmaceutica Sinica B, as much as 90% of clinical drug development fails.
CombinAble uses AI to “learn” from data and physics-based principals to optimize antibody sequences in one step. Developed with input from AION Labs' pharmaceutical partners and publicly available data, its platform represents a shift from trial-and-error to a streamlined, computer-simulated solution for more effective therapeutics.
"Our emphasis is on combining various advanced machine learning models, including large protein language models, molecular dynamics simulations and multi-objective optimization algorithms, integrating them into a single unified solution for targeted antibody discovery," said Daria Kokh, co-founder and CEO of CombinAble.AI. "AION Labs' resources and expertise have been pivotal in shaping our company, providing the material and logistical support to help us refine our scientific strategy and expand the team. As we progress, the knowledge and insights provided by its pharmaceutical partners will be invaluable for the development and commercialization of our solution."
Last month, AION Labs announced the formation of TenAces Biosciences, which will leverage machine learning for molecular glue discovery. That company also aims to improve the success rate of the discovery of treatments for a range of conditions.