cellVIEW: a Tool for Illustrative and Multi-Scale Rendering of Large Biomolecular Datasets.

Area of Science:

• Biophysics
• Computational Biology
• Scientific Visualization

Background:

• Visualizing large biomolecular datasets at the atomic level presents significant computational and graphical challenges.
• Existing tools often struggle with the scale and complexity of structures like viruses and bacterial organisms.

Purpose of the Study:

• To introduce cellVIEW, a novel system for interactive, atomic-level visualization of large biomolecular datasets.
• To enable domain experts to model and visualize structures containing billions of atoms in real-time.

Main Methods:

• Integration of acceleration techniques for real-time graphics performance (60 Hz).
• Implementation of a level-of-detail (LOD) scheme to enhance rendering speed and reduce visual clutter.
• Extension of rendering methods for dynamic GPU generation of DNA strands from control points.
• Development within a game engine to leverage advanced graphics capabilities and reduce development workload.

Main Results:

• Achieved real-time graphics performance for datasets of large viruses and bacterial organisms.
• Successfully reduced visual clutter and accelerated rendering through the LOD scheme.
• Enabled dynamic, GPU-accelerated generation of nucleic acid strands.
• Provided a freely available, extensible tool for biomolecular visualization.

Conclusions:

• cellVIEW offers a unique solution for interactive visualization of large biomolecular landscapes at the atomic level.
• The system's integration with a game engine makes advanced graphics techniques accessible.
• cellVIEW is currently the only tool capable of handling such large-scale, atomic-level biomolecular visualization interactively.