Related Experiment Video
Updated: Jan 10, 2026
14:04
Detection of Protein Aggregation using Fluorescence Correlation Spectroscopy
Published on: April 25, 2021
6.0K
Aggregation of Intrinsically Fluorescent Proteins: Combining Coarse-Grained Molecular Dynamics with Alphafold
Margherita Bini1,2, Giorgia Brancolini3, Valentina Tozzini2,1,4
1Scuola Normale Superiore, Piazza San Silvestro, 12, Pisa 56127, Italy.
ACS Omega
|November 24, 2025
Summary
This study compares molecular dynamics simulations and AlphaFold3 (AF3) for predicting protein aggregation. Results guide force field selection and integration of AI with simulations for complex protein systems.
Area of Science:
- Protein aggregation
- Computational biophysics
- Artificial intelligence in structural biology
Background:
- Protein aggregation is common in both disordered and folded proteins.
- It is influenced by hydrophobic and electrostatic forces.
- Molecular dynamics (MD) simulations, often low-resolution, are typically used to study aggregation.
Purpose of the Study:
- Compare coarse-grained MD simulations with AlphaFold3 (AF3) for predicting protein aggregation.
- Assess the performance of different force fields in MD simulations.
- Evaluate AF3's capability in predicting protein complex structures relevant to aggregation.
Main Methods:
- Coarse-grained molecular dynamics simulations.
- Utilized four variants of intrinsically fluorescent proteins.
- Employed AlphaFold3 (AF3) for structure prediction.
Main Results:
- Performance comparison between MD simulations with various force fields and AF3.
- Insights into the accuracy of different computational approaches for protein aggregation prediction.
- Demonstrated the potential of integrating AF3 with MD simulations.
Conclusions:
- Results aid in selecting or designing force fields for systems with disordered and folded proteins.
- Provides a framework for combining AlphaFold predictions with MD simulations.
- Highlights the evolving role of AI in understanding protein aggregation.
