Data Mining and Computational Modeling of High-Throughput Screening Datasets.

Area of Science:

• Cheminformatics
• Drug Discovery Informatics
• Computational Chemistry

Background:

• Significant investments in high-throughput screening (HTS) have led to large structure-activity datasets available in public databases.
• The growth of academic drug discovery centers necessitates advanced informatics tools for data analysis and knowledge extraction.
• Existing tools may not fully address the need for integrated data mining, visualization, and modeling of HTS data.

Purpose of the Study:

• To introduce novel web-based data mining and visualization tools within the CDD Vault platform for high-throughput drug discovery.
• To describe CDD Models for sharing and creating predictive models from distributed, heterogeneous datasets.
• To demonstrate the utility of these tools using public HTS datasets.

Main Methods:

• Development of a web-based module utilizing a modern reactive design technology stack.
• Integration of data mining and visualization capabilities directly within the CDD Vault platform.
• Implementation of CDD Models for collaborative model building and prediction sharing.

Main Results:

• The CDD Vault platform now offers real-time manipulation and visualization of thousands of molecules.
• Researchers can effectively mine, share, and learn from large HTS datasets.
• CDD Models facilitate the creation and sharing of predictive models from diverse data sources.

Conclusions:

• The presented CDD Vault tools provide powerful capabilities for analyzing and modeling high-throughput screening data.
• These informatics solutions support early-stage drug discovery in academic settings.
• The platform complements existing cheminformatics tools, enhancing data mining and modeling approaches for HTS data.