Analytical Approaches to Improve Accuracy in Solving the Protein Topology Problem

Area of Science:

• Structural biology
• Computational biology
• Biophysics

Background:

• Determining 3D structures of biological macromolecules is crucial for genomics and proteomics.
• Cryo-electron microscopy (cryo-EM) offers a promising method, but resolution limitations often prevent direct atomic structure determination.
• Secondary structure information is detectable via cryo-EM, enabling computational modeling approaches like de novo modeling.

Purpose of the Study:

• To enhance the accuracy of de novo modeling for macromolecular structures derived from cryo-electron microscopy data.
• To improve the mapping between detected secondary structures and the amino acid sequence using computational tools.
• To refine the DP-TOSS (Dynamic Programming for determining the Topology Of Secondary Structures) algorithm for more precise topology determination.

Main Methods:

• Proposed modifications to the scoring methodology of the DP-TOSS tool.
• Implemented and tested three novel scoring schemes: skeleton-based, geometry-based analytical, and multi-well potential energy-based functions.
• Evaluated the improved DP-TOSS performance on a dataset of 25 proteins.

Main Results:

• The modified scoring schemes demonstrated an improvement in the accuracy of DP-TOSS.
• A combination of the proposed scoring functions yielded the best performance in solving the topology determination problem.
• The enhanced DP-TOSS successfully mapped detected secondary structures to amino acid sequences for macromolecular proteins.

Conclusions:

• Modifications to the DP-TOSS scoring methodology significantly enhance its accuracy for macromolecular structure analysis.
• The integration of novel scoring functions provides a more robust approach to de novo modeling using cryo-EM data.
• This work advances computational methods for determining protein structures, crucial for understanding biological functions.