Characterizing the Sequence Landscape of Peptide Fibrillization with a Bottom-Up Coarse-Grained Model.

Area of Science:

• Computational biology
• Biophysics
• Molecular modeling

Background:

• Amyloid aggregation is central to proteinopathies, but its complex timescales pose computational challenges.
• Current coarse-grained models often combine top-down and bottom-up parameterization methods.

Purpose of the Study:

• To present a predictive, sequence-transferable, bottom-up coarse-grained model for peptide aggregation.
• To enable accurate simulation of amyloid fibril nucleation and growth.

Main Methods:

• Systematic development using atomistic simulation data.
• Application of extended-ensemble relative entropy minimization for parameterization.
• Coupling coarse-grained simulations with a genetic algorithm.

Main Results:

• The model accurately recovers peptide conformational properties using a reduced amino acid alphabet.
• It predicts secondary structures from peptide sequences.
• It successfully simulates experimentally characterized amyloidogenic peptide aggregation.

Conclusions:

• The developed model provides a powerful tool for studying amyloid aggregation.
• It facilitates the identification of sequence features promoting thermodynamically favorable and kinetically accessible fibrillar states.