GalaxyDomDock: An Ab Initio Domain-domain Docking Web Server for Multi-domain Protein Structure Prediction

Area of Science:

• Computational Biology
• Structural Bioinformatics
• Protein Structure Prediction

Background:

• Many proteins consist of multiple domains, and their relative orientations are crucial for function.
• Predicting multi-domain protein structures is challenging when individual domain structures are known, but their orientations are not.

Purpose of the Study:

• To develop and evaluate GalaxyDomDock, an ab initio domain-domain docking method for predicting multi-domain protein structures.
• To provide insights into protein function by modeling different domain orientations.

Main Methods:

• GalaxyDomDock utilizes an ab initio approach based on the protein-protein docking method GalaxyTongDock.
• It employs a geometric filter with the Dijkstra algorithm to remove infeasible domain orientations.
• Loop sampling method FALC is used for predicting domain linker conformations.

Main Results:

• GalaxyDomDock demonstrated superior performance compared to existing methods like AIDA and Rosetta on benchmark datasets.
• It showed comparable or better results than AIDA, even for proteins with discontinuous domains and AlphaFold-predicted domains.

Conclusions:

• GalaxyDomDock is an effective tool for predicting domain-domain docking and multi-domain protein structures.
• The method advances the field of structural bioinformatics and aids in functional genomics research.
• The GalaxyDomDock web server is publicly accessible for research use.