The density-based many-body expansion for poly-peptides and proteins.

Area of Science:

• Quantum Chemistry
• Computational Biology
• Biophysics

Background:

• Accurate quantum-chemical calculations are crucial for understanding large biomolecular systems.
• Fragmentation schemes are essential for treating complex systems efficiently.
• Developing accurate machine-learning potentials for proteins requires reliable computational methods.

Purpose of the Study:

• To present a novel fragment-based method for quantum-chemical treatment of proteins.
• To reduce fragmentation errors in total energies of polypeptides and proteins.
• To extend the applicability of density-based many-body expansion (db-MBE) to biomolecular systems.

Main Methods:

• A fragment-based method utilizing calculations of single amino acids and their dimers.
• Combining the molecular fractionation with conjugate caps (MFCC) scheme with density-based many-body expansion (db-MBE).
• Two-body extension of the MFCC scheme.

Main Results:

• Achieved reduction of fragmentation error in total energies to approximately 1 kJ mol^-1 per amino acid.
• Demonstrated accuracy across various structural motifs in polypeptides and proteins.
• Successfully extended the db-MBE applicability from molecular clusters to proteins.

Conclusions:

• The presented fragment-based method offers an efficient and accurate approach for quantum-chemical calculations of large biomolecules.
• The combination of MFCC and db-MBE significantly minimizes fragmentation errors.
• This method provides a strong foundation for developing accurate machine-learning potentials for proteins.