TomoSim: Simulation of Filamentous Cryo-Electron Tomograms

Area of Science:

• Structural biology
• Biophysics
• Computational biology

Background:

• Automated filament tracing in cryo-electron tomography (cryo-ET) requires robust validation methods.
• Experimental cryo-ET data often suffers from low resolution, noise, and missing wedge artifacts, obscuring true filament structures.
• A reliable ground truth is essential for accurately testing and improving tracing algorithms.

Purpose of the Study:

• To present TomoSim, a novel software tool for simulating realistic cryo-electron tomography maps.
• To enable the creation of simulated tomographic data with known ground truth filament traces.
• To facilitate the validation of automated filament tracing algorithms in cryo-electron tomography.

Main Methods:

• TomoSim generates simulated tomographic maps based on a provided filament trace.
• The tool automatically matches simulation parameters to experimental cryo-electron tomography map characteristics.
• Key simulation features include Fourier space wedge masking, noise color simulation, and signal-to-noise ratio matching.

Main Results:

• The developed prototype successfully simulates tomographic maps with realistic artifacts.
• The simulation parameters can be automatically adjusted to match experimental data.
• The software provides a controllable environment for testing tracing algorithms.

Conclusions:

• TomoSim offers a valuable solution for generating ground truth data for cryo-ET.
• This tool is crucial for the reliable validation and development of automated cytoskeletal filament tracing methods.
• Future applications include enhancing the accuracy and reliability of structural biology analyses using cryo-ET.