Using dynamics-based comparisons to predict nucleic acid binding sites in proteins: an application to OB-fold domains

Area of Science:

• Biophysics
• Computational Biology
• Structural Biology

Background:

• Proteins can be aligned using dynamical properties, not just sequence or structure.
• Dynamics-based alignments are effective for comparing dissimilar protein families.
• This study focuses on predicting nucleic acid-binding regions using protein dynamics.

Purpose of the Study:

• To develop and validate a method for predicting protein regions involved in nucleic acid binding.
• To utilize dynamics-based alignments for functional site prediction.
• To apply the method to the OB-fold family, specifically the AXH-domain.

Main Methods:

• Utilizing dynamics-based alignments to compare protein families.
• Correlating statistically significant alignments with known nucleic acid binding regions.
• Applying the validated method to predict binding sites in the AXH-domain.

Main Results:

• Dynamics-based alignments successfully identified regions involved in nucleic acid binding within the OB-fold family.
• The method predicted putatively functional nucleic acid binding regions in the AXH-domain.
• The approach demonstrates a promising general strategy for predicting functional regions based on dynamics.

Conclusions:

• Dynamics-based protein alignment is a powerful tool for predicting functional regions, particularly for nucleic acid binding.
• This method offers a novel approach for characterizing protein function in unstudied families like the AXH-domain.
• The validated scheme provides a generalizable strategy for large-scale functional site prediction in proteins.