3D Visualization of Molecular Data using a Cloud Streaming Framework for Web-based Large Model Viewing

Merry Shiyu Wang, Michael Zyracki, Andrew Kimoto, Malte Tinnus, Florencio Mazzoldi

3D visualization of large and multi-scale biological data, from macro-molecular structures to whole organisms, is integral to building predictive models for biomedical research. However, existing industry standard desktop, plug-in, and web-based applications can easily push beyond the limits of advanced in-core processing and rendering. Autodesk Research Molecular Viewer is designed to overcome limitations of scalability, capability, accessibility, collaboration, and outreach for molecular datasets. Layered on an extensible cloud-based visualization framework optimized for large 3D models and data, the Viewer demonstrates common usage scenarios of molecule interaction by visualizing RCSB Protein Data Bank files directly in the web browser. The technology leverages a scalable, cloud-based translation and streaming pipeline, and utilizes three.js on webGL-enabled browsers to support a distributed data model. A root manifest streams in smaller component parts, which render as required, to absolve the need for entire model storage in memory. The Molecular Viewer visualization framework can be extended to other 3D data, enabling exploration of large-scale biomedical models.

BioVis 2015 Information