MCCE2: improving protein pKa calculations with extensive side chain rotamer sampling.

Area of Science:

• Biophysical Chemistry
• Computational Biology
• Protein Biochemistry

Background:

• Accurate prediction of protein and ligand pK(a)s is crucial for understanding biochemical processes.
• Traditional methods often simplify side chain flexibility and dielectric environments.
• Multiconformation continuum electrostatics (MCCE) offers a more nuanced approach.

Purpose of the Study:

• To describe and validate the MCCE2 methodology for calculating pK(a)s.
• To assess the impact of conformational sampling and optimization on pK(a) prediction accuracy.
• To investigate the influence of protein structure source and dielectric constants on results.

Main Methods:

• Utilized Monte Carlo simulations within the MCCE framework to explore conformational space.
• Employed QUICK and FULL calculation protocols with varying degrees of side chain optimization.
• Tested calculations against a dataset of 305 experimental pK(a) values from 33 proteins, including bacteriorhodopsin.

Main Results:

• QUICK calculations achieved an RMSD of 1.34, with 84% of errors <1.5 pH units.
• FULL calculations improved accuracy to an RMSD of 0.90, with 90% of errors <1.5 pH units.
• Calculations using X-ray structures yielded better pK(a) predictions than NMR structures; a dielectric constant of 4 performed comparably to 8.

Conclusions:

• MCCE2 provides a robust and accurate method for predicting pK(a)s, especially when incorporating detailed conformational sampling and optimization.
• The choice of input structure (X-ray vs. NMR) significantly impacts prediction accuracy.
• The MCCE2 program is available for download, facilitating further research in protein electrostatics.