PRA-Pred: Structure-based prediction of protein-RNA binding affinity.

Area of Science:

• Biochemistry
• Structural Biology
• Bioinformatics

Background:

• Protein-RNA complexes are crucial for cellular functions.
• Understanding their binding affinity is key to deciphering recognition mechanisms.

Purpose of the Study:

• To identify factors influencing protein-RNA binding affinity.
• To develop a predictive model for binding affinity.
• To create a user-friendly web server for predictions.

Main Methods:

• Compiled binding affinity data (ΔG) for 217 protein-RNA complexes.
• Extracted structure-based features including RNA parameters, interaction energies, and atomic contacts.
• Developed multiple regression models and validated using jack-knife tests.

Main Results:

• RNA base-step parameters, interaction energies, and atomic contacts significantly impact binding affinity.
• Models achieved a correlation of 0.77 between experimental and predicted binding affinities.
• Introduced the PRA-Pred web server for predicting protein-RNA binding affinity.

Conclusions:

• The study provides insights into protein-RNA recognition mechanisms.
• The developed models and web server can aid in investigating these interactions.
• This resource has potential applications in developing therapeutic strategies.