Reconstruction of Atomistic Structures from Coarse-Grained Models for Protein-DNA Complexes

Area of Science:

• Computational Biology
• Structural Biology
• Biomolecular Modeling

Background:

• Coarse-grained (CG) simulations accelerate sampling of large biomolecular complexes but lack atomic detail.
• Reconstructing all-atom (AA) structures from CG models is crucial for accurate analysis and refinement.

Purpose of the Study:

• To develop an efficient method for reconstructing AA protein-DNA complex structures from CG models.
• To accurately model the protein-DNA interface during AA structure reconstruction.

Main Methods:

• Developed a novel method to reconstruct atomic DNA structures from a three-site per nucleotide CG model using a DNA fragment library.
• Modeled protein-DNA interfaces by referencing known side chain orientations and utilized existing tools for other regions.
• Validated the protocol through self-reproduction, base pair reproducibility, and analysis of atomic contacts.

Main Results:

• Successfully reconstructed atomic details of DNA structures from CG models.
• Accurately modeled protein-DNA interfaces, preserving key atomic contacts.
• Demonstrated the feasibility of subsequent all-atom simulations with the reconstructed structures.

Conclusions:

• The developed protocol provides an efficient and accurate approach for converting CG protein-DNA models to AA structures.
• This method enhances the utility of CG simulations by enabling high-fidelity downstream all-atom modeling and analysis.