DEMO-EMol

DEMO-EMol integrates deep learning-based map segmentation and chain fitting to accurately assemble protein-nucleic acid complex structures from cryo-EM density maps.

Key Features:

• Deep learning-based map segmentation: Segments protein and nucleic acid regions from cryo-EM density maps using deep learning.
• Chain fitting and assembly: Fits protein and nucleic acid chain models into their respective segmented maps to assemble the overall complex.
• Domain-level fitting and optimization: Performs domain-level fitting and optimization for protein chains.
• Quality assessment: Provides overall and residue-level quality assessments of the final assembled complex model.

Scientific Applications:

• Protein-nucleic acid complex modeling: Assembles protein-nucleic acid complex structures from cryo-EM density maps and independently modeled chain structures.
• Protein-protein complex modeling: Demonstrates superior performance for protein-protein complex modeling compared to state-of-the-art methods.
• Cryo-EM structure determination: Supports atomic structure modeling across a broad resolution range from 1.96 to 12.77 Å.

Methodology:

DEMO-EMol first segments protein and nucleic acid regions from the density map using deep learning. The overall complex is then assembled by fitting protein and nucleic acid chain models into their respective segmented maps, followed by domain-level fitting and optimization for protein chains.