CS-MINER: A Tool for Association Mining in Binding-Database.

Area of Science:

• Computational chemistry
• Cheminformatics
• Drug discovery

Background:

• Understanding the relationship between chemical space and biological targets is crucial for drug discovery.
• Existing tools lack systematic correlation between medicinal target space and chemical space.

Purpose of the Study:

• To introduce CS-MINER (Chemical Space Miner), a novel software tool for the visualization and analysis of drug-like chemical space.
• To systematically correlate medicinal target space with chemical space using a comprehensive molecular database.

Main Methods:

• Compilation of a large molecule database (110 biological targets from Binding-DB).
• Calculation of 1497 physicochemical properties for each molecule.
• Application of discriminant analysis and multivariate classifiers for data analysis and virtual screening.

Main Results:

• CS-MINER effectively separates molecules based on therapeutic targets and activities.
• Multivariate classifiers enable ligand-based virtual screening of over 0.5 million molecules from PubChem and ZINC.
• Validation of models by comparing selected subspaces with DrugBank molecules.

Conclusions:

• CS-MINER provides an interactive environment for visualizing chemical subspaces in 2D and 3D plots.
• The software facilitates comparison of active biosimilar molecules within chemical space.
• CS-MINER is a valuable, freely available tool for drug discovery research.