CASA: an efficient automated assignment of protein mainchain NMR data using an ordered tree search algorithm.

Area of Science:

• Biochemistry and Structural Biology
• Computational Biology
• Nuclear Magnetic Resonance Spectroscopy

Background:

• Accurate protein structure, interaction, and dynamics analysis relies on rapid and automated assignment of 3D protein backbone triple-resonance NMR spectra.
• Existing methods may lack speed or flexibility in handling diverse NMR datasets.

Purpose of the Study:

• To introduce CASA, a novel computational tool for automated assignment of triple-resonance NMR spectra.
• To demonstrate the speed and accuracy of CASA across various protein sizes and data types.

Main Methods:

• Development of a depth-first ordered tree search algorithm named CASA.
• Utilizing hand-edited peak-pick lists from a flexible number of triple resonance experiments.
• Testing on simulated peak lists for proteins up to 723 residues and experimental data from four proteins.

Main Results:

• CASA generated accurate assignments comparable to literature values within minutes of CPU time for both simulated and experimental data.
• The program successfully assigned spectra for proteins up to 723 residues.
• CASA demonstrated robustness when tested against datasets analyzed by other methods and under various challenging conditions.

Conclusions:

• CASA provides a fast and automated solution for assigning 3D protein backbone triple-resonance NMR spectra.
• The software is effective for proteins of significant size and diverse experimental conditions.
• CASA represents a valuable advancement for structural biology research requiring rapid protein analysis.