Modeling protein structure at near atomic resolutions with Gorgon

Area of Science:

• Structural Biology
• Biophysics
• Computational Biology

Background:

• Electron cryo-microscopy (cryo-EM) is crucial for studying macromolecular assemblies in near-native conditions.
• Cryo-EM often yields non-atomic resolution maps, requiring computational methods for detailed structural analysis.
• Recent advances enable near-atomic resolution cryo-EM maps, facilitating manual model construction.

Purpose of the Study:

• To introduce Gorgon, an interactive modeling toolkit for intermediate to near-atomic resolution density maps (10-3.5 Å).
• To provide computational tools for de novo protein backbone model generation and 3D volume annotation.
• To demonstrate Gorgon's utility in analyzing cryo-EM data.

Main Methods:

• Gorgon employs a de novo modeling procedure combining sequence-based secondary structure prediction.
• It integrates feature detection and geometric modeling techniques for initial protein backbone generation.
• The toolkit serves as an extensible interactive visualization platform with diverse computational tools.

Main Results:

• Gorgon successfully generates initial protein backbone models from intermediate-resolution density maps.
• The toolkit facilitates the annotation of various 3D volumes derived from cryo-EM data.
• Applicability was demonstrated using cryo-EM maps of Rotavirus and Rice Dwarf Virus.

Conclusions:

• Gorgon is an effective tool for modeling protein structures from intermediate to near-atomic resolution cryo-EM maps.
• It streamlines the process of structural analysis and annotation for large macromolecular complexes.
• The toolkit enhances the interpretation of cryo-EM data, contributing to a deeper understanding of molecular function.