The Three-D Informatics Group at the Lister Hill National Center for Biomedical Communications is developing software for the registration, segmentation, analysis and visualization of 3D data from confocal microscopy, transmission electron tomography, and high resolution dual-beam scanning electron microscopy. These technologies are exploring cell biology at nanometer scales. The group crafts individual models, visualizations, and presentations of the processed data from these modalities of microscopy for publication and interactive viewing among our colleagues in the biological sciences. We are conducting this research in partnership with the National Cancer Institute (NCI) Laboratory for Cell Biology. As forms of data acquisition, these processes resemble the very successful Visible Human Project (VHP), where cryosectioning, a destructive imaging technique, was used to capture a representation of human anatomy at unprecedented levels of quality and fidelity. Effectively, a human subject was sliced at 1/3 mm intervals and the exposed sections photographed digitally to yield a human anatomical study with pixel resolutions as fine as 350 microns. In a similar approach, dual-beam scanning electron microscopy starts with a fixed cellular sample and ablates the cell using a focused ion beam, slicing the sample at intervals as fine as 5 nanometers. The exposed sections are imaged using a scanning electron microscope. The comparable approaches of these imaging techniques allow the Three-D Informatics Group to apply tools similar to those used on VHP data for registration, segmentation, and image analysis. The flourishing VHP Insight Toolkit (ITK) open source software initiative, sponsored by NLM, is used extensively in this project. We also employ public tools such as 3D Slicer, a software system that uses ITK and is produced by Harvards National Alliance for Medical Image Computing, one of the NIH Roadmap National Centers for Biomedical Computing funded through the National Institute of Biomedical Imaging and Bioengineering. Since FY2009, the group has provided software tools to NCIs Laboratory of Cell Biology to promote high throughput microscopy and improve the understanding of cellular biology, particularly in the communication of HIV AIDS among dendritic cells, macrophages, and T-Cells. Illuminations created by the team have aided in communicating these results and their importance to the community. These results have been published in journals including the Journal of Structural Biology. One image that features cell membrane models generated through this software was featured on the cover of the November-December 2009 IAVI Report (the publication on AIDS Vaccine Research), volume 13, number 6, with the subtitle Visualizing HIV. Other publications include a study of the transfer of HIV to T cells, described in the Proceedings of the National Academy of Sciences. NCI acquired a new microscope in FY2011 that produces images with improved resolution, greater contrast, and greater relative signal to noise properties. During FY2012, the group helped develop software tools to analyze the differentiation of murine myocytes as they develop from cultured mouse adult stem cells to muscle tissue, work that has continued throughout FY 2013. Members of the team have been investigating new computing methods for performing high-performance, multi-threaded, rapid calculations on graphics processing units for sub-volume averaging to achieve super-resolution in transmission electron tomography. This technique in 3D high-resolution electron microscopy is unlocking the structure of proteins in HIV virions. In addition to software development, the team created visualizations, animations, and 3D models of protein spikes from the surface of HIV virions as part of the scientific review of this research program conducted by NCI. The team has been continuing to develop these methods to extend them to study unconstrained microscopy problems known as single particle reconstruction with some early indications that we could accelerate the computing in this area as well.

Project Start
Project End
Budget Start
Budget End
Support Year
5
Fiscal Year
2014
Total Cost
Indirect Cost
Name
National Library of Medicine
Department
Type
DUNS #
City
State
Country
Zip Code