Computational estimation of specific side chain interaction energies in alpha helices

Area of Science:

• Computational biology
• Protein structure prediction
• Biophysics

Background:

• Accurate prediction of protein secondary structure, particularly alpha helices, is crucial for understanding protein function.
• Existing helix/coil transition algorithms rely on parameter sets for amino acid interactions, which can be improved with theoretical energy calculations.

Purpose of the Study:

• To theoretically estimate specific side chain-side chain interaction energies within alpha helices using a structure energy-based program (Perla).
• To substitute these computed energies into the AGADIR algorithm for predicting peptide helical content.
• To experimentally validate computed interaction energies for specific amino acid pairs.

Main Methods:

• Utilized the Perla computer program for structure energy-based protein design to calculate side chain-side chain interaction energies in alpha helices.
• Integrated computed interaction energies into the AGADIR helix/coil transition algorithm.
• Experimentally determined interaction energies for Lys-Ile, Thr-Ile, and Phe-Ile amino acid pairs at i,i + 4 positions.

Main Results:

• Peptide helical content predictions using computed energies showed high correlation (0.91) with experimental data, comparable to original parameters.
• Experimental validation confirmed favorable agreement with computed theoretical estimates for Lys-Ile, Thr-Ile, and Phe-Ile interactions.
• Computed energies for Thr-Ile and Phe-Ile interactions outperformed those based on chemical similarity.

Conclusions:

• Structure energy-based computational methods, like Perla, can accurately predict amino acid pairwise interaction energies in alpha helices.
• These computational techniques enhance the accuracy of protein structure prediction and sequence design tools.
• The findings support the development of advanced computational tools for protein engineering and structural biology.