expertise

Finding a new chemistry that targets a specific receptor or biomolecular target is not a new science. Just as there are a myriad of ways to select a starting point, select a target, identify tool compounds, find hit classes, perform lead optimization on these classes and advance chemistry, the in vitro to in vivo paradigm is not nearly as fast as computational approaches for hit identification and lead optimization. From a known structure, either apo/holo forms can be used or explored directly, or as homology models for known orthologs of interest and automated procedures for generating structural hypotheses (pharmacophores). Using in silico libraries of millions of lead-like compounds that are in-stock or rapidly synthesizable, one can now triage chemistries using a combination of filters, classifier models, pharmacophores, molecular docking, that allows us to explore a million compounds an hour on a standard desktop. Imagine the power of such a streamlined approach on cloud infrastructures to be able to more efficiently identify and nominate vendor-ready collections? Contact me to learn more.

Any chemical that is designed for a specific target may also, inadvertently, hit multiple other targets within the same species (i.e. human drug, human side-effect) or different species (human drug, ecological side-effect). Knowing how a unique chemical structure relates to its biological activity and its fate within the body, the "off-target" binding sites and various receptors and enzymes, is a critical aspect of computational discovery and is one of the leading causes of attrition during early drug discovery. Similarly, the propensity for a given chemical nomination to be advanced in a drug discovery pipeline is defined by a gauntlet of safety questions related not only to on/off-target toxicity of the chemical, but also safety on beneficials, pharmacokinetics or ADME, formulation and matrix compatibility, cost-of-goods considerations, synthesizability, and stability (i.e. photostability, thermal stability, oxidative degradation) of both the parent hit/lead compound, and its metabolic progeny. In this aspect, computational de-risking can be used to reduce off-target liabilities of a hit-class early on in the discovery pipeline to reduce downstream costly attrition. Contact me to learn more.

Getting a handle on the competitive intellectual property (IP) landscape and freedom to operate (FTO) can be a daunting undertaking for any discovery molecular scientist. Knowing what chemistry is out there, who owns it, where there are structural loopholes / liabilities in a patent and determining just how "novel" your chemistry is can be challenging. Even trying to strip a patent down (i.e. markush structures) can be overwhelming, and trying to make sense of the data in chemical-biological claims in a patent isn't a job easy for everyone. Luckily, there are opportunities for intelligently navigating the IP landscape and developing either novel "composition of matter" or "methods of use" patents that can strengthen your IP strategy and portfolio. Contact me to learn more.