Accurate secondary structure prediction and fold recognition for circular dichroism spectroscopy

Area of Science:

• Biophysics
• Structural Biology
• Spectroscopy

Background:

• Circular dichroism (CD) spectroscopy is vital for protein structure analysis.
• Existing algorithms struggle with diverse beta-structures, impacting accuracy for mixed or beta-rich proteins.
• Spectral diversity in beta-structures has limited CD spectroscopy's reliability for unusual protein types.

Purpose of the Study:

• To address limitations in CD spectral analysis of protein secondary structures.
• To develop a method accounting for beta-sheet orientation and twist for improved accuracy.
• To enable reliable protein fold prediction from CD spectra.

Main Methods:

• Developed the beta-structure selection (BeStSel) method.
• Incorporated parallel/antiparallel orientation and twisting of beta-sheets into analysis.
• Applied BeStSel to conventional CD and synchrotron radiation CD (SRCD) spectroscopy data.

Main Results:

• BeStSel accurately estimates secondary structure for a wide range of proteins.
• The method reliably distinguishes parallel and antiparallel beta-sheets.
• Protein fold topology (CATH classification) can be predicted from single CD spectra.
• Investigated amyloid fibril structures using BeStSel and SRCD.

Conclusions:

• BeStSel significantly enhances the reliability of CD spectroscopy for protein secondary structure estimation.
• The method offers a valuable tool for analyzing proteins, especially membrane proteins, aggregates, and fibrils.
• A web server provides accessible, reliable structure analysis for researchers, complementing X-ray and NMR techniques.