Gell: A GPU-powered 3D hybrid simulator for large-scale multicellular system.

Area of Science:

• Computational Biology
• Biophysics
• Systems Biology

Background:

• Hybrid computational models simulate multicellular dynamics but face computational cost limitations for large-scale systems.
• Increasing biological system complexity necessitates efficient simulation methods for accessible hardware.

Purpose of the Study:

• To develop a fast and memory-efficient GPU-based platform, Gell (GPU Cell), for large-scale hybrid computational modeling.
• To enable complex multicellular system simulations on standard computational devices.

Main Methods:

• Developed Gell, an open-source GPU-accelerated hybrid computational modeling platform.
• Implemented full parallelization on GPU for enhanced computational efficiency.
• Introduced a novel voxel sorting method to optimize cell-cell mechanical interaction modeling.

Main Results:

• Gell efficiently simulates systems with tens of millions of cells on personal computers.
• Demonstrated ~150X acceleration compared to state-of-the-art CPU-based simulators.
• Achieved one-tenth of the memory requirement compared to CPU-based methods.

Conclusions:

• Gell significantly reduces computational cost for large-scale hybrid simulations.
• The platform facilitates advanced research in complex multicellular systems, including disease modeling like ductal carcinoma in situ (DCIS).
• Gell democratizes access to powerful computational biology tools.