A unique approach for protein secondary structure comparison under TOPS representation

Area of Science:

• Structural bioinformatics
• Computational biology
• Biophysics

Background:

• Evaluating protein secondary structure similarity is crucial for understanding protein function.
• Existing methods for comparing protein secondary structure elements (SSE) have limitations, necessitating new approaches.

Purpose of the Study:

• To introduce a novel alignment-free method for assessing protein secondary structure similarity using Topology of Protein Structure (TOPS) representations.
• To develop a new approach for constructing phylogenetic trees based on protein secondary structure comparisons.

Main Methods:

• A 2D numerical representation of SSE is created, with points weighted by frequency.
• Moments-of-inertia are calculated using weighted coordinate means.
• A four-component descriptor is generated, and Manhattan distance is used to create a distance matrix for phylogenetic tree construction (NJ method).

Main Results:

• The proposed TOPS-based method was applied to 36 SSE elements from the Chew-Kedem database.
• Phylogenetic trees were generated and compared against established methods (Clustal-Omega, LZ-Complexity, SED, TOPS+TOC).
• The new method demonstrated superior performance in constructing phylogenetic trees for SSE comparison.

Conclusions:

• The developed alignment-free method using TOPS representations provides an effective means for protein secondary structure comparison.
• This approach advances the analysis of protein structure similarity and aids in phylogenetic tree construction.
• The method offers a valuable alternative for researchers studying the relationship between protein structure and function.