An associative memory Hamiltonian model for DNA and nucleosomes

Area of Science:

• Computational Biology
• Biophysics
• Molecular Modeling

Background:

• Understanding chromatin structure is crucial for gene regulation.
• Existing models often lack the ability to capture both DNA mechanics and higher-order chromatin organization.
• Simulating DNA at a single base level up to complex structures presents a significant challenge.

Purpose of the Study:

• To introduce a novel computational model, the Widely Editable Chromatin Model (WEChroM), for simulating DNA and nucleosomes.
• To enable the study of chromosomes from single base pairs to higher-order chromatin structures.
• To develop a model that accurately reproduces DNA's mechanical properties and behavior under supercoiling.

Main Methods:

• Development of the WEChroM Hamiltonian, incorporating chain connectivity, steric interactions, and associative memory terms.
• Simulation of circular DNA behavior under positive and negative supercoiling.
• Application of the model to study DNA unwrapping from nucleosomes.
• Implementation within the OpenMM simulation toolkit.

Main Results:

• WEChroM successfully reproduces DNA's bending and twisting persistence lengths, including temperature dependence.
• The model accurately recapitulates the formation of plectonemes and structural defects in response to supercoiling.
• WEChroM exhibits asymmetric behavior under positive and negative supercoiling, consistent with experimental observations.
• The model's associative memory component effectively reproduces the free energy of partial DNA unwrapping from nucleosomes.

Conclusions:

• WEChroM provides a versatile and accurate platform for simulating DNA and chromatin structures at multiple scales.
• The model's ability to capture complex mechanical properties and behaviors makes it suitable for investigating gene regulation and chromatin dynamics.
• WEChroM's open-source availability facilitates broader research in molecular systems and structural ensembles of genes.