Galux, an artificial intelligence (AI)-based new drug development corporations, said on the 26th that by using its platform GaluxDesign, it has secured antibody candidates with drug-level binding strength with only 50 designs per target.
Galux explained that this achievement shows that, unlike conventional antibody discovery methods, an era has opened in which the required functions can be designed from the start and secured in a short time.
Galux has its own platform, GaluxDesign, which combines AI with physicochemical principles. Its core technology is a de novo method that designs new proteins from scratch without referencing existing proteins. Galux was founded by Seok Cha-ok, a professor in the Department of Chemistry at Seoul National University who has researched protein structure prediction technology for more than 20 years.
Traditional antibody discovery relies on experimental methods such as animal immune responses or large-scale library screening, making it difficult to predict the desired antibodies. Even after antibodies are secured, multiple optimization steps such as improving immunogenicity and enhancing affinity are required, so the discovery process usually takes more than a year.
To overcome these limitations, Galux designed and validated antibodies with strong binding to multiple targets through AI-based antibody design technology. It confirmed that the AI-predicted structures matched the experimental structures from cryo-electron microscopy (Cryo-EM). Without broad exploration, it also achieved results in which more than 30% bound precisely to therapeutic targets with only 50 designs.
According to the company, the results represent performance improved by thousands of times compared with not only traditional experimental methods but also existing AI approaches, and rank among the top within the five reported groups of de novo antibody design cases to date. The company said many of these showed binding strength at a level that could be developed into drug candidates within a month even without separate optimization.
Galux CEO Seok Cha-ok said, "This study shows a transition from an era of 'discovery' of antibodies to an era of 'design' of the required antibodies from the start," adding, "Based on this, we will fully take on the challenge of developing differentiated antibody therapeutics that were difficult to secure with existing methods."