This project brings together a strong interdisciplinary team of investigators to develop powerful image information processing and database tools for the exponentially growing bio-molecular image data, thus enabling a new generation of bio-image informatics. Contemporary microscopic techniques, including immunofluorescence microscopy, electron microscopy, GFP visualization and atomic force microscopy have ;ontributed enormously to our current understanding of molecular cell biology in its many forms. Individual research labs often generate hundreds or even thousands of images weekly. Unfortunately, most of these images are analyzed, labeled and archived manually. Further, the vast majority of these images never get published, despite the fact that many of them would be valuable for other researchers. The problems inherent in manual labeling and archiving, together with the lack of a central and searchable repository for images of all sorts (analogous to DNA sequence databases) is a major impediment to progress in essentially all fields under the general description of molecular and cellular biology. These problems also represent a major impediment to progress in the emerging area of bio-image informatics, and tools for organizing and information processing of such data are urgently needed. By coupling recent advances in image processing, pattern recognition and data mining, with the enormous amount of data that is being generated in bio-molecular imaging, significant progress can be made in our understanding of various cellular and subcelluar processes.
