Procedure for Transferable Coarse-Grained Models of Aqueous Polysaccharides

Area of Science:

• Computational Chemistry
• Materials Science
• Biophysics

Background:

• Developing accurate coarse-grained (CG) models for polysaccharide solutions is crucial for understanding their behavior in various applications.
• Existing CG models often struggle with transferability across different molecular weights and concentrations, limiting their predictive power.
• Atomistic Molecular Dynamics (MD) simulations provide detailed insights but are computationally expensive for large systems.

Purpose of the Study:

• To present a novel hybrid procedure for generating transferable CG models of aqueous polysaccharide solutions.
• To overcome limitations of previous CG modeling approaches, such as inaccurate aggregation behavior and end-to-end distances.
• To enable the application of CG models to long polysaccharides and varying concentrations.

Main Methods:

• A hybrid approach combining Boltzmann Inversion (BI) and the Multiscale Coarse-Graining (MS-CG) method was employed.
• A separation-ansatz and explicit 1-3 and 1-4 nonbonded intramolecular interactions were utilized to improve model accuracy.
• The procedure was validated for transferability across different degrees of polymerization and concentrations, and extended to implicit solvent models and various force fields.

Main Results:

• The developed CG models demonstrate transferability over varying polysaccharide concentrations and degrees of polymerization.
• The CG models accurately capture the behavior of long polysaccharides, addressing previous discrepancies with atomistic simulations.
• A transferable implicit solvent model was generated, compatible with different atomistic force fields.
• Application to hemicellulose polysaccharides revealed that branching significantly increases water uptake capacity compared to linear structures.

Conclusions:

• The hybrid BI/MS-CG procedure provides a robust and transferable method for CG model development of polysaccharide solutions.
• The improved CG models are suitable for simulating long polysaccharides and can be extended to implicit solvent representations.
• The findings on hemicellulose water uptake highlight the importance of molecular structure in determining solution properties and align with experimental observations.