i3Drefine software for protein 3D structure refinement and its assessment in CASP10

Area of Science:

• Computational Biology
• Structural Biology
• Biophysics

Background:

• Protein structure refinement aims to improve the accuracy of modeled protein structures.
• Refinement has gained importance in Critical Assessment of techniques for Protein Structure prediction (CASP) experiments.
• Despite increased participation, consistent quality improvement in protein structure refinement remains a challenge.

Purpose of the Study:

• To develop an automated and efficient protein 3D structure refinement method.
• To evaluate the performance of the developed method against state-of-the-art techniques in a blind test setting.

Main Methods:

• Developed i3Drefine, an automated protein structure refinement tool.
• Employed an iterative energy minimization algorithm with composite force fields and hydrogen bonding network optimization.
• Utilized a powerful all-atom composite physics and knowledge-based force fields.

Main Results:

• i3Drefine ranked as the best server method in the CASP10 refinement category.
• Demonstrated consistent improvement in both global and local structural quality metrics compared to initial structures.
• Outperformed other state-of-the-art refinement methods in CASP10.

Conclusions:

• i3Drefine is a highly effective, fully-automated protein structure refinement server.
• The method offers consistent quality improvements for protein models.
• Free access to i3Drefine software is provided to the scientific community.