ALLO: A tool to discriminate and prioritize allosteric pockets

Area of Science:

• Biochemistry and Structural Biology
• Computational Chemistry
• Drug Discovery

Background:

• Allosteric proteins represent a significant portion of human drug targets.
• Targeting allosteric sites requires identifying these pockets and understanding their function.

Purpose of the Study:

• To characterize allosteric pockets using physicochemical descriptors.
• To develop predictive models for distinguishing allosteric from orthosteric pockets.
• To create models for prioritizing allosteric pockets on a given protein.

Main Methods:

• Characterization of allosteric pockets using physicochemical descriptors.
• Comparison of allosteric pockets with surface protein pockets.
• Training predictive models for pocket classification and prioritization.

Main Results:

• Physicochemical descriptors effectively differentiate allosteric pockets.
• Developed models can discriminate between allosteric and orthosteric pockets.
• Models can prioritize potential allosteric sites within a protein.

Conclusions:

• The developed models can aid in identifying novel allosteric sites.
• This approach may facilitate the discovery of new allosteric drug targets.
• Associated datasets and a Python program are publicly available for use.