Recognition of functional sites in protein structures

Area of Science:

• Computational biology
• Structural bioinformatics
• Drug discovery

Background:

• Identifying similar binding sites on proteins is essential for predicting molecular interactions and classifying protein functions.
• Existing methods often rely on sequence or structural similarities, limiting their scope.

Purpose of the Study:

• To introduce SiteEngine, a novel method for recognizing similar protein binding sites irrespective of sequence or fold similarity.
• To enable functional site recognition, prediction of molecular interactions, and aid in drug design.

Main Methods:

• Utilizes a low-resolution surface representation based on chemically important points.
• Employs hashing of physico-chemical properties and hierarchical scoring for efficient similarity exploration.
• Applies the method to search known sites, compare potential sites, and discover unknown functional sites within protein structures.

Main Results:

• Demonstrates high efficiency and speed in recognizing similar binding sites.
• Successfully applied to large-scale searches of the Protein Data Bank.
• Identified potential secondary drug binding sites and novel targets for drug design.

Conclusions:

• SiteEngine is a robust and efficient tool for functional site recognition and protein classification.
• The method facilitates understanding drug side effects and discovering new therapeutic targets.
• Enables accurate functional predictions and classification of binding patterns.