Plant & Animal Genomes XIII Conference

January 15-19, 2005
Town & Country Convention Center
San Diego, CA

Development Of Genome Analysis Tool Using Scanning Near-Field Optical / Atomic Force Microscopy

Yoshitaka Suetsugu¹ , Tomoyuki Yoshino² , Seigo Kuwazaki¹ , Hirokazu Takahashi¹ , Junko Narukawa¹ , Motoharu Shichiri² , Daisuke Fukushi³ , Kazuei Mita¹ , Shigeru Sugiyama² , Toshio Ohtani² , Kimiko Yamamoto¹

¹  National Institute of Agrobiological Sciences, 1-2 Owashi, Tsukuba, Ibaraki 305-8634, Japan

²  National Food Research Institute, 2-1-12 Kannondai, Tsukuba, Ibaraki 305-8642, Japan

³  Niigata University , 1-757 Asahimachi-dori, Niigata 951-8510, Japan

Scanning Probe Microscopy (SPM) represented by Atomic Force Microscopy (AFM) and Scanning Near-field Optical/Atomic Force Microscopy (SNOM/AFM) has nanometer-scale resolution and does not require special treatment, such as staining and metal-coating. Recently, our group launched a project called “Development of SPM genome analysis method” to apply SPM to a genome analysis. In this project, development of a technology to automatically map clones onto a chromosome with high resolution is one of our goals. First of all, algorithms for predicting chromosomal volume and for locating signals of fluorescence in situ hybridization (FISH) were contrived and these were implemented on the graphical user interface (GUI) application.
The outline of algorithms is as follows. In predicting chromosomal volume, at first, the topological data obtained by SNOM/AFM were transformed into the image by quantization. Then, filtering was applied to the image. The filtered image was segmented by a thresholding, and the resultant binary image was labeled using 4-neighbor algorithm. The chromosomal volume was calculated by the information of labeled region and corresponding height. In locating FISH signal, the signal region was detected by filtering and thresholding of the raw fluorescence image. Then, template-matching between the topographic image and fluorescence image was performed to map FISH signal onto chromosome.
These algorithms are useful for high-resolution mapping of clones onto chromosomes and also for a development of a novel system to construct physical map by SPM.
This research is supported by fund of Promotion of Basic Research Activities for Innovative Biosciences from BRAIN.