Careers
Laboratory Integration
The iterative laboratory ↔ computation process tests many more molecules in less time than the synthesis ↔ test optimization cycle. Instead, with Verseon's computational engine, scientists get detailed and accurate SAR on a large number of compounds families, which they then use to identify the best potential avenues for optimization. Accurate and rapid computation helps them in exploring their hypotheses in much greater detail and in reaching the desired endpoint more consistently, as opposed to abandoning two-thirds of the discovery projects after years of work.
Computational Modeling Engine
Some of the factors that underlie Verseon's proprietary computational modeling engine are:
- Advancements in accurate modeling of the fundamental physical processes that contribute to the binding free energy of a compound-target complex in water.
- Numerical algorithms that are capable of calculating the outcome of such processes accurately and rapidly.
- Application of above algorithms to a vast chemical space of hundreds of millions of compounds, and identifying the best compounds for synthesis, testing, and optimization.
We specifically eschew rules-based and machine-learning methods that attempt to statistically extrapolate solutions from databases of successful drugs. Such shortcuts are easier to implement than a rigorous physics- and chemistry-based methodology, but have proven to be of narrow utility and are so unreliable as not to form the heart of an industrial-strength drug discovery pipeline.
At Verseon, we capture the essential physics that defines the interaction between a drug-like molecule and a protein target residing in an aqueous medium. Unprecedented accuracy in capturing the physics has enabled us to identify the best compounds for any target from large libraries generated by our Molecule Design Engine.
Molecule Design Engine
The molecules are generated from drug-like building blocks using a reaction knowledge base that ensures each resulting molecule is unique and, very importantly, can be synthesized using known chemistry.The pool of molecules can be wide, spanning a large number of scaffolds if medicinal chemists are looking for novel hits for a target, or it can be narrow, spanning fewer scaffolds but generating many more variations to help medicinal chemists do rapid optimization.
The engine provides medicinal chemists with the facility to generate only those molecules that satisfy a range of properties to suit particular criteria, such as solubility or molecular weight or those that contain a certain set of chemical groups.
Verseon has used the molecule design engine to generate tens of millions of novel compounds for its drug development programs. To put the capabilities of this engine in perspective, the compound collections available to large pharmaceutical companies are typically no larger than two million compounds, and those two million compounds are restricted to variations around a few thousand distinct scaffolds. The limitations of these corporate compound libraries is one reason that fully two thirds of drug discovery projects industry wide are yielding no useful drug compound hits.