Ion Mobility Mass Spectrometry Guided Modeling with AlphaFold and Rosetta Improves Protein Complex Structure...

Area of Science:

• Structural biology
• Computational biology
• Biophysics

Background:

• Ion mobility mass spectrometry (IM-MS) offers insights into protein size and shape via collision cross section (CCS), but lacks atomic resolution.
• AlphaFold excels at monomeric protein structure prediction but faces challenges in modeling protein complexes accurately.

Purpose of the Study:

• To develop an integrated computational approach combining IM-MS data with AlphaFold and Rosetta for improved protein complex structure prediction.
• To enhance the accuracy of modeling protein assemblies by leveraging experimental structural constraints.

Main Methods:

• Integration of experimental collision cross section (CCS) data from IM-MS.
• Utilized AlphaFold for subunit structure prediction and Rosetta for complex assembly via a docking pipeline.
• Developed a novel scoring function to evaluate the quality of predicted protein complexes.

Main Results:

• Improved root mean square deviation (RMSD) values for 68% (26 of 38) of tested protein complexes compared to AlphaFold-Multimer.
• Achieved significant accuracy improvements, with 16 complexes transitioning from >4 Å RMSD to <4 Å RMSD.
• Demonstrated a robust strategy for enhancing protein complex modeling.

Conclusions:

• The integrated IM-MS, AlphaFold, and Rosetta approach effectively refines protein complex structure prediction.
• This method provides a powerful solution for overcoming limitations in modeling the assembly of protein complexes.
• The strategy offers a significant advancement in computational structural biology for complex systems.