Data-driven Design

Nowadays, high-throughput screening has been shown to have an irreplaceable role in drug discovery. However, low assay throughput or high screening costs can prevent screening a large number of compounds. Therefore, a design of compound sets using data-driven can improve the predictability of active compounds in HTS. In addition, with the growing interest in the importance of training set composition for predicting active substances, more and more data-driven based strategies are used to cover the descriptor space.

BOC Sciences has employed two data-driven methods: chemoinformatics-based designand chemical reactions-based design methods to construct predictive models for selecting compounds from the screening set for subsequent screening.

Figure 1. Diverse libraries compared with focused libraries. (Shardul, P. et al. 2018)

Library design for targets with few known active chemotypes assays

Investigation of relevant chemical space is important for targets with few known active chemotypes or phenotypic assays, and our groups have adopted a diversity-based library design method by optimizing biological relevance and compound diversity to provide multiple starting points for further development:

• Chemical descriptors are used to characterize compounds based on structure and/or physicochemical properties

A wide range of chemical descriptors such as fingerprint-based, shape-based or pharmacophore-based are available

• Biological descriptors are employed to represent compound phenotypic effects and biological activity against the drug proteome

BOC Sciences supports various biological descriptors such as affinity fingerprints or high-throughput screening fingerprints, HTS-FP

Library design for targets with many known active chemotypes

In general, focused libraries are designed for well-studied targets such as GPCRs, kinases, ion channels, etc

• Diversity-based screening is performed to discover active chemotypes
• Focused libraries center around active chemotypes are selected from larger diversity-based libraries using structure-based and/or ligand-centric similarity metrics
• Develop ligands with desired properties
• Fully consider both the diversity of compounds and the inclusion of known active chemotypes

Figure 2. Overview of recent studies improving (scaffold) hit rates and providing insights into compound mode of action. (Shardul, P. et al. 2018)

BOC Sciences provides professional, rapid and high-quality services of Data-driven Design design at competitive prices for global customers. Personalized and customized services of Data-driven Design design can satisfy any innovative scientific study demands. Our clients have direct access to our staff and prompt feedback to their inquiries. If you are interested in our services, please contact us immediately!