Voronoi Kinase Library DB

Voronoi has built a library of kinase-selective compounds. The biggest strength of Voronoi’s library is the database measuring pharmacokinetics and profiling the full panel of 468 kinases’ activity for each compound. Thanks to this, when a particular molecule target is proven by the academia to be correlated to a disease, Voronoi can search its own DB to find a compound that can selectively target that kinase. This allows Voronoi to develop candidate compounds without delay. We currently have a kinase profiling DB for more than 3,000 compounds, and 1,000 new compounds are added each year. This broad DB profiling kinase selectivity of compound structure will build on, to boost Voronoi’s productivity over time.

A.I. Prediction

Voronoi has an AI platform that quickly identifies candidates using compounds searched in the in-house DB. The AI platform can analyze relationship between the structure of compounds and particular kinase activity, and enable hybridization of the core structure units of the compound to rapidly create multiple, more suitable derivative compounds.

CADD(Computer Aided Drug Design)

Molecular modeling using computers is a core technology applicable to all areas of new drug development. We have built a platform that allows us to select early hit compounds through in silico screening, utilizing a large in-house compound DB. Using in silico technology for identifying early hit for a new target can dramatically save research time and cost compared to conventional drug discovery research. Also, since we already know the binding pattern of the disease target and the inhibitor, we can expedite efficacy optimization and lead discovery.

As GPU cards allow large volume of calculation these days, the activity of newly designed drugs can be accurately predicted. We have GPU cluster equipment that enables us to predict the activity of new compounds quantitatively, which maximizes research efficiency. When GPU is used to calculate molecular dynamics, the binding free energy of the inhibitor and the target protein can be predicted with accuracy. This allows us to develop candidate-stage compounds quickly, after testing fewer compounds.

The in silico technology allows us to not only design compounds but also predict pharmacological aspects. We predict BBB penetration rate, which is essential for CNS diseases, internal absorption and metabolite, and apply the result to actual medicine synthesis. We have the capabilities and technology to discover pre-clinical candidates within 1 or 2 years, utilizing the in silico platform technology.

Kinase platform