Secondary structure assignment that accurately reflects physical and evolutionary characteristics.

Area of Science:

• Biochemistry
• Structural Biology
• Computational Biology

Background:

• Secondary structure is crucial for protein classification, feature identification, and structure prediction.
• Existing methods for secondary structure assignment often rely on hydrogen bonding, limiting their accuracy and scope.
• Current methods have differing definitions, particularly for secondary structure termini, and cannot identify all structural classes.

Purpose of the Study:

• To develop a more accurate method for assigning protein secondary structure.
• To overcome limitations of existing hydrogen-bonding-dependent methods.
• To enable comprehensive secondary structure assignment, including isolated strands and poly-proline helices.

Main Methods:

• Developed a novel secondary structure assignment method based on main chain geometric parameters.
• Defined helices by fitting residues within a cylinder and classifying helix types (alpha, 3(10), pi) by peptide bond angles.
• Assigned beta-strands using backbone dihedral angles and alternating peptide bonds, independent of hydrogen bonding.

Main Results:

• The new method demonstrates improved accuracy in secondary structure assignment compared to existing approaches.
• Observed stronger amino-acid trends at helix caps and reduced concatenation of secondary structural elements.
• Achieved enhanced secondary structure-guided sequence alignment, reflecting better physical and evolutionary characteristics.

Conclusions:

• The developed geometric method provides a more comprehensive and accurate assignment of protein secondary structure.
• Assignments using this method better represent the physical and evolutionary properties of proteins.
• The associated software is publicly available for use in protein structure analysis.