We scan new podcasts and send you the top 5 insights daily.
Unlike antagonist therapeutics that rely on tight binding to block targets, agonist antibodies require lower affinity and rapid off-rates to signal continuously without receptor quenching. Applying machine learning based on natural ligands prevents teams from wasting cycles engineering inappropriate high affinities.
The relationship between a multi-specific antibody's design and its function is often non-intuitive. LabGenius's ML platform excels by exploring this complex "fitness landscape" without human bias, identifying high-performing molecules that a rational designer would deem too unconventional or "crazy."
The industry's focus on antibodies, which are easy to generate, may be a case of technology dictating the science. Dr. Radvanyi argues that natural ligand-receptor interactions, which often rely on lower affinity and higher avidity, could offer a more nuanced and effective way to modulate immune pathways than high-affinity agonist antibodies.
Contrary to the popular belief that antibody development is a bespoke craft, modern methods enable a reproducible, systematic engineering process. This allows for predictable creation of antibodies with specific properties, such as matching affinity for human and animal targets, a feat once considered a "flight of fancy."
Unlike antibodies with flexible loops allowing for induced fit, de novo designed proteins are hyper-rigid. This pre-organized structure leads to rapid binding (fast on-rates) but also makes them susceptible to rapid unbinding (fast off-rates), as the rigid interface is more easily displaced by solvent.
A key advantage of LabGenius's AI platform is its unbiased approach, which proposes multi-specific antibody designs that traditional engineers might dismiss as too complex or unmanufacturable. By testing these counter-intuitive candidates, the platform identifies high-performing molecules that would otherwise be overlooked.
Traditional antibody optimization is a slow, iterative process of improving one property at a time, taking 1-3 years. By using high-throughput data to train machine learning models, companies like A-AlphaBio can now simultaneously optimize for multiple characteristics like affinity, stability, and developability in a single three-month process.
For complex biologics with many binders, chasing astronomical affinity is counterproductive and risks off-tumor toxicity. A better strategy is to use binders with modest affinity and leverage the massive avidity gained from multiple binding sites. This provides a 'finer dial' to tune specificity and improve the therapeutic window.
Many innovative drug designs fail because they are difficult to manufacture. LabGenius's ML platform avoids this by simultaneously optimizing for both biological function (e.g., potency) and "developability." This allows them to explore unconventional molecular designs without hitting a production wall later.
Traditional drug design crystallizes a receptor to understand its structure, removing it from its biological context. Metaphor reverses this by first studying a receptor's dynamic interactions in living systems, ensuring its antibodies are functionally active from the start.
Dr. Radvanyi explains that immune agonist drugs often fail because accelerating a biological pathway is inherently less controllable than inhibiting one. This is analogous to genetic knockouts being more straightforward than over-expression models, presenting a core challenge in drug development beyond just finding the right target.