DUKE INVENTOR: Bruce Donald, Jonathan Jou, Marcel Frenkel, Mark Hallen
Gavilán Biodesign combines state of the art physics-based modeling with our unique high accuracy AI platform to computationally screen trillions of molecules in order to find therapeutics that can overcome resistance. Gavilán is the first to add the dimension of evolution and time in our screens! We do this through our proprietary design software suite called Sylph that can design both biologics and small molecules for specificity and resistance resilience.
Our algorithms are based on rigorous mathematical guarantees allowing us to make confident predictions about our chemical space and continuously improve our predictions through improvements to our entropy-aware, continuously flexible, high accuracy biophysical models that are already the most sophisticated models of their kind for combinatorial drug design.
Explore Beyond, Leave Nothing Unseen
Gavilán Biodesign uses its proprietary computational chemistry approach to develop novel antineoplastic therapeutics that are resilient to the emergence of drug resistance. The ability of cancer to develop resistance to therapeutics is the number one contributing force to mortality. Gavilan is building on the success of the OSPREY software package developed in the Donald lab at Duke University, expanding it to not only predict but also design new molecules capable of overcoming drug resistance.
Gavilán is a results-oriented company that believes that accuracy of predictions is paramount. Therefore, Gavilán utilizes state-of-the-art algorithms that are guaranteed to find the best results within our chemical space despite our screens routinely having trillions of molecules. Nothing is left unseen: our algorithms guarantee no good compound is left behind and no resistance mutation gets through. As our accuracy is only limited by our physical models, Gavilán builds the most sophisticated models of any group performing combinatorial computational chemistry by taking into account entropy and continuous flexibility.