Creating the next generation of protein-based drugs and enzymes powered by machine learning
What is the Challenge?
80% of drug targets are complex protein interfaces beyond the reach of traditional drugs. Small molecules are too small to successfully bind the targets, and biologics are too large to get inside the cells where most of those challenging targets reside. Peptides have long been proposed as a "Goldilocks" molecule with just the right size to perpetuate the cell membrane and bind to the target, but accessing that chemical space has proven very challenging. Recent successes (e.g. Merck’s PCSK9 and JnJ’s IL-23 inhibitors) have foreshadowed a next-generation of peptide macrocycles with membrane-permeability capabilities that could be extended to other challenging drug targets; however, identifying and optimizing these molecules experimentally remains very challenging and can take several years. What if there was a way to use AI to create a shortcut to the right chemical space?
What is the Solution?
Menten AI, a pioneering company in the field of generative AI for drug design, is making significant strides in the pharmaceutical industry. Supported by leading venture capital firms such as Uncork Capital, Khosla Ventures, and Y-Combinator, Menten AI has developed a groundbreaking platform capable of designing advanced macrocyclic therapeutics. This platform stands out as the first of its kind, proficient in crafting potent and membrane-permeable macrocycles in just days. The robustness of their platform is evident in its application to intricate drug targets, including protein-protein interfaces and intracellular targets. What sets Menten AI apart is their successful collaborations with top-10 pharmaceutical companies, which have greatly accelerated their preclinical drug discovery efforts.
The technology behind Menten AI's success integrates cutting-edge machine learning techniques with physics-based models and quantum chemistry simulations. This fusion not only speeds up the design process but also optimizes the macrocycles to possess desired drug-like properties, such as high potency and membrane permeability, within a remarkably short timeframe. Unlike traditional methods that involve screening billions of molecules, Menten AI's platform employs generative AI to construct potential macrocycles from scratch, effectively pinpointing the right chemical space. The platform's efficacy has been proven through various research programs, leading to the discovery of potent macrocycles for complex drug targets. Notably, it has even generated antiviral peptides against COVID-19. Currently, Menten AI is leveraging this innovative platform in partnership with leading pharmaceutical companies, aiming to expedite their drug development programs.