Leveraging protein dynamics to identify cancer mutational hotspots using 3D structures

Area of Science:

• Genomics
• Structural Biology
• Computational Biology

Background:

• Cancer driver gene identification relies on recurrence-based methods and 3D protein structures.
• Existing methods often neglect protein dynamics, a crucial factor in protein function.

Purpose of the Study:

• To develop a novel framework for identifying cancer driver genes by incorporating protein dynamics into mutational hotspot analysis.
• To improve the sensitivity and accuracy of cancer driver detection.

Main Methods:

• Developed a framework to identify mutational hotspot communities in protein structures using dynamics-based residue networks.
• Mapped mutations to protein structures and weighted residue contacts by correlated motions.
• Applied the method to the TCGA PanCancer Atlas missense mutation catalog to detect signals of positive selection.

Main Results:

• Predicted mutational hotspots in 434 genes, with significant enrichment in tumor progression-associated biological processes.
• Demonstrated that incorporating protein dynamics significantly enhances the sensitivity of driver detection compared to existing methods.
• Identified novel insights into cancer-driving mechanisms through a dynamics-based approach.

Conclusions:

• The proposed dynamics-based framework effectively identifies cancer driver genes by revealing mutation clusters within functionally relevant protein communities.
• This approach offers a more sensitive and comprehensive method for cancer driver discovery, advancing the field of cancer genomics.
• Understanding protein dynamics is critical for accurate identification of cancer-associated mutational hotspots.