γ-TEMPy: Simultaneous Fitting of Components in 3D-EM Maps of Their Assembly Using a Genetic Algorithm

Area of Science:

• Structural Biology
• Computational Biology
• Biophysics

Background:

• Macromolecular complexes are essential for cellular functions.
• Determining the structure of these complexes is crucial for understanding their mechanisms.
• 3D-electron microscopy (3D-EM) provides low-resolution density maps of macromolecular complexes.

Purpose of the Study:

• To develop a computational method for building accurate models of macromolecular complexes.
• To utilize 3D-electron microscopy density maps and atomic structures of components for complex reconstruction.
• To assess the method's performance across various resolutions and experimental conditions.

Main Methods:

• Development of a genetic algorithm for macromolecular complex assembly.
• Use of vector quantization for efficient sampling of map feature points.
• Implementation of a fitness function combining mutual information and clash penalty scores.

Main Results:

• The genetic algorithm correctly identified the topology of macromolecular assemblies in 90% of cases with simulated 10 Å resolution maps.
• Performance decreased with lower resolution maps (70% at 15 Å, 60% at 20 Å).
• The method successfully identified the correct topology for all four tested assemblies with experimental maps ranging from 7.2 to 23.5 Å.

Conclusions:

• The developed genetic algorithm is effective for reconstructing macromolecular complexes from 3D-EM data.
• Map feature-point quality significantly impacts assembly fitting accuracy, especially without additional experimental data.
• This method provides a valuable tool for structural biologists and computational chemists.