SI2008 Multiscale Image Analysis and Visualization
This track will cover multiscale image acquisition, tomographic reconstruction, meshing, visualization, and basic model building techniques for mesoscale simulations.
Time: 9 am to 12 pm
- Tuesday, Aug 5, High Throughput Electron Tomography through the use of Imaging Portals and Data Repositories
- Instructor: James Obayashi, Masako Terada
- Users will learn about basic tomography principles for producing high quality 3D tomograms, including program suites such as IMOD and TxBR. The Telescience Imaging Portal and methods for stream lining the processing steps will be presented. In addition, the Cell Centered Database which can be used to house large tomographic datasets will be introduced. Afterwards, a hands on tutorial using the IMOD tomography suite will be covered.
- Participants will be provided with a computing workstation, but bringing one's laptop, with the IMOD suite installed, is encouraged. The software package and installation directions are found at http://bio3d.colorado.edu/imod/
- Wednsday, Aug 6, Mesoscale imaging, processing, and modeling of microdomains in cardiac myocytes
- Instructor: Zeyun Yu, Masahiko Hoshijima
- Users will be introduced to animal models for heart failure and arrhythmia, and 3D tomographic reconstruction of the dyadic cleft and t-tubular network. A number of image processing, feature extraction, and mesh generation will be presented to build spatially realistic models using GAMer and other tools.
- Thursday, Aug 7, Multiscale visualization using PMV
- Instructor: Ruth Huey, Sargis Dallakyan
- The Python Molecule Viewing environment (PMV) is a freely available molecular viewer designed by Michel F. Sanner and developed in the Molecular Graphics Lab. (MGL). Besides the traditional features found in other tools such as Rasmol, PMV implements a number of unique and innovative features.
- Users will be introduced to the Python Molecular Viewer, and its extended capabilities using Vision in visualizing large scale datasets such as viral capsids, and complex meshes. Attendees will also learn how to use PMV for visualizing molecules in various ways and produce publication-quality images. The commands covered include loading molecules and displaying them using various representations, computing molecular surfaces and secondary structure representations, selecting subsets of atoms for coloring and labeling and using APBS to compute and visualize electrostatic potentials. No prior knowledge of Python is required.
- Friday, Aug 8, Multiscale Image Analysis Workflow Management using Vision
- Instructors: Guillaume Vareille, Luca Clementi
- Please note: this session is a joint session with Track 1.
- Users will learn how to use the Vision workflow and visual programming environment to build workflows for molecular visualization, image processing, and coupling different applications into a computational pipeline through the development of Vision nodes. Uses will also learn how to develop complex workflows using remote Opal based web services for distributing long running jobs to remote resources including the TeraGrid. How to develop Vision nodes based on local applications or remote web services.