From Atoms to Cells: AI-Based Structure Prediction Fueling Molecular Dynamics Simulations in Computational Structural...

Area of Science:

• Computational Biology
• Biophysics
• Systems Biology

Background:

• Biological system simulations have advanced from single proteins to entire cells.
• Traditional molecular dynamics (MD) simulations required experimentally determined structures.
• Artificial intelligence (AI) models like AlphaFold now predict protein structures rapidly and accurately.

Purpose of the Study:

• To explore the convergence of AI and MD simulations for whole-cell modeling.
• To bridge the gap between static structural models and dynamic cellular processes.
• To highlight the potential of digital cells in biological research and drug discovery.

Main Methods:

• Utilizing AI-powered protein structure prediction (e.g., AlphaFold).
• Employing molecular dynamics (MD) simulations to capture molecular motion and function.
• Integrating structural and dynamic data for comprehensive cellular simulations.

Main Results:

• AI enables accurate protein structure prediction, expanding simulation scope.
• MD simulations reveal dynamic conformational changes and mechanisms.
• The combination facilitates the simulation of complex biological assemblies and cellular processes.

Conclusions:

• AI and MD simulations are key to achieving whole-cell simulations with high resolution.
• This approach transforms biological research, drug discovery, and synthetic biology.
• Digital cells are emerging as fundamental tools for scientific exploration.