Can Simple Interaction Models Explain Sequence-Dependent Effects in Peptide Homodimerization?

Area of Science:

• Biophysics
• Computational Chemistry
• Materials Science

Background:

• Predicting peptide behavior is vital for amyloid disease research, therapeutic stability, and biomaterial design.
• Molecular simulations offer atomic-level insights beyond experimental databases for model development.
• Understanding peptide aggregation and self-assembly is critical across multiple scientific disciplines.

Purpose of the Study:

• To develop and validate simple, computationally inexpensive models for predicting peptide homodimerization.
• To assess the accuracy of simplified interaction models against detailed all-atom simulations.
• To identify key physical drivers of peptide association for improved predictive modeling.

Main Methods:

• Utilized all-atom molecular simulations to calculate reference dimerization free energy profiles and binding constants.
• Developed and statistically assessed simplified interaction models based on scaling laws.
• Investigated serine-glycine peptides with varying hydrophobic leucine mutations as a case study.

Main Results:

• Certain combinations of simple scaling laws accurately reproduced detailed all-atom simulation results.
• A phenomenological hydrophobic force law demonstrated high explanatory power for peptide association.
• Coarse measures of entropic effects in binding were found to be physically relevant to peptide association.

Conclusions:

• Simplified models can effectively capture complex peptide interaction and aggregation behaviors.
• Hydrophobic interactions and entropic effects are critical driving forces in peptide self-assembly.
• This approach facilitates the development of rapid, accurate methods for peptide behavior prediction.