Improving All-Atom Molecular Dynamics Models for Quantitative Prediction of Nanopore Blockade Current

Area of Science:

• Computational biology
• Biophysics
• Nanotechnology

Background:

• Nanopore sensors detect biomolecules via ionic current changes.
• Relating molecular structure to current modulations requires computational methods.
• All-atom molecular dynamics (MD) is crucial for nanopore sensor development.

Purpose of the Study:

• To evaluate standard MD force fields for simulating single-stranded DNA in nanopores.
• To improve the accuracy of MD simulations for nanopore sensing applications.
• To refine force field parameters for better agreement with experimental data.

Main Methods:

• All-atom molecular dynamics simulations of single-stranded DNA passing through a protein nanopore.
• Testing CHARMM36, AMBER Parmbsc1, and DES-AMBER force fields.
• Experimental validation and refinement of non-bonded interactions in the CHARMM36 force field.

Main Results:

• Standard MD force fields (CHARMM36, AMBER Parmbsc1, DES-AMBER) failed to accurately reproduce experimental nanopore blockade currents.
• Surgical corrections to non-bonded interactions in CHARMM36 significantly improved simulation accuracy.
• The refined CHARMM36 force field achieved favorable agreement between simulation and experimental data.

Conclusions:

• Existing MD force fields require refinement for accurate nanopore DNA sensing simulations.
• Targeted adjustments to non-bonded interactions enhance the predictive power of MD simulations.
• This refined force field has potential applications in DNA-protein interaction studies.