Assessing the quality of modelled 3D protein structures using the ModFOLD server

Area of Science:

• Structural biology
• Computational biology
• Bioinformatics

Background:

• Model quality assessment programs (MQAPs) are essential for evaluating the accuracy of 3D protein structure models.
• Quality scores, both global and per-residue, are vital for determining the usability of protein models in downstream applications.
• The Critical Assessment of Techniques for Protein Structure Prediction (CASP) competition significantly drives advancements in MQAPs.

Purpose of the Study:

• To review the role of MQAPs in protein structure prediction.
• To discuss prominent MQAPs from past CASP competitions.
• To introduce and provide usage guidelines for the ModFOLD and IntFOLD-QA tools.

Main Methods:

• Review of existing literature and CASP competition outcomes.
• Description of downloadable and webserver-based MQAP tools: ModFOLD3, ModFOLDclust, ModFOLDclustQ, ModFOLDclust2, and IntFOLD-QA.
• Development of practical, step-by-step usage guides for the presented tools.

Main Results:

• Overview of key MQAPs and their performance in CASP.
• Availability of user-friendly downloadable and webserver tools for model quality assessment.
• Demonstration of tool application in improving tertiary structure prediction, ligand binding site prediction, and oligomer predictions.

Conclusions:

• MQAPs are indispensable for assessing 3D protein model quality.
• The ModFOLD and IntFOLD-QA tools offer practical solutions for quality assessment and model improvement.
• These tools facilitate diverse applications, including structure prediction refinement and biological interaction studies.