Evaluation of a grid based molecular dynamics approach for polypeptide simulations

Area of Science:

• Biomedical Research
• Computational Biology
• Biophysics

Background:

• Molecular dynamics (MD) simulations are crucial for understanding biological macromolecule behavior in silico.
• Traditional MD simulations are computationally intensive, requiring significant time and resources for large biomolecules like proteins.
• High-performance computing is essential for obtaining biologically relevant data from MD simulations.

Purpose of the Study:

• To develop a distributed computing infrastructure for running molecular dynamics simulations on a grid platform.
• To enable parallel submission of numerous short simulations that collectively yield significant biological insights.
• To enhance the efficiency and accessibility of MD simulations for biomedical research.

Main Methods:

• Implementation of a grid-based infrastructure for distributed molecular dynamics simulations.
• Parallel submission of multiple independent, short simulation trajectories.
• Job chaining within simulations to mitigate data loss and manage data transfer sizes on the grid.

Main Results:

• The developed infrastructure successfully supports distributed molecular dynamics simulations.
• The parallel execution of short trajectories provides valuable biological information.
• The system demonstrates high scalability, confirming the suitability of grid computing for MD.

Conclusions:

• Distributed computing on grid platforms is a highly scalable and effective approach for molecular dynamics simulations.
• The implemented infrastructure enhances the feasibility of conducting complex MD studies in biomedical research.
• This method offers a powerful alternative for generating significant biological data more efficiently.