Introduction to the Workshop
URLs Provided by Attendees

Abstracts

Mapping

Informatics

Sequencing

Instrumentation

Ethical, Legal, and Social Issues

Infrastructure

The electronic form of this document may be cited in the following style:
Human Genome Program, U.S. Department of Energy, DOE Human Genome Program Contractor-Grantee Workshop IV, 1994.

Abstracts scanned from text submitted for November 1994 DOE Human Genome Program Contractor-Grantee Workshop. Inaccuracies have not been corrected.

CGH Imaging for Routine Use

Damir Sudar[1,5], Jim Mullikin[1], Steve Lockett[1], Jim Piper[3,1], Gus van der Feltz[4,1], David Kaszuba[2], Mark de Kanter[4,1], Dan Pinkel[2,1], Joe Gray[2,1]
[1]Resource for Molecular Cytogenetics; MS 74-157; Lawrence Berkeley Laboratory; 1 Cyclotron Road; Berkeley, CA 94720. [2]University of California, San Francisco, CA. [3]Medical Research Council, Edinburgh, UK. [4]Delft University of Technology, Delft, The Netherlands. [5]Corresponding author

The Resource for Molecular Cytogenetics at LBL/UCSF uses CGH analysis for research into many different tumors and cell types potentially containing genetic abnormalities. In a typical CGH application test and reference genomic DNA, each labeled with a different fluorochrome, are hybridized to normal metaphase chromosomes. The ratio of intensity of the 2 fluorochromes along the length of the chromosome is proportional to the local DNA copy number. Around 25 researchers routinely perform CGH experiments which puts special requirements on the systems used for the analysis. The typical rate of image acquisition is 100 Mbytes/day of compressed data on 2 digital imaging microscopes; 5 workstations are used for analysis and image data is maintained on multiple servers and storage systems. We describe the CGH imaging procedures in use at the Resource and their implementation for routine use.

Data is obtained from CGH experiments using a digital imaging microscope (QUIPS) which we developed. QUIPS emphasizes straightforward acquisition of multi-color fluorescence images of high quality. User-friendly programs have been developed for convenient and rapid acquisition of the required images which are automatically stored on the fileserver in a single dataset. Fully automated routines take care of migration of older datasets to a tape robot based storage system to make space for new data. Image data stays on the local fileserver for an average of 4 months which gives the researcher time to analyze it. Data archived on the tape robot system can be recalled overnight.

Images stored on the fileserver are recalled on one of the analysis workstations using programs under SCILimage. Individual fluorochrome images are read in from the dataset for visual analysis and for semi-automated processing according to Piper et al. [1]. Visual analysis consists of inspection of the individual fluorochrome images and multi-color compositions. Semi-automated processing begins with chromosome segmentation based on the counterstain and reference DNA images after background correction and is followed by a cluster detection and decomposition step [2]. Normalizers for each DNA image are calculated from the derived segmentation mask. Chromosome identification is done visually after image enhancement of the counterstain image. We are working on the integration of computer-assisted karyotyping techniques from standard cytogenetics to facilitate the identification of chromosomes. For each chromosome a local background is calculated and a profile is extracted for both hybridization images by integrating the background corrected and normalized pixel values along slices orthogonal to the chromosome axis. A division of the 2 profiles yields the CGH ratio profile for this chromosome.

The ratio profiles for multiple chromosomes are averaged to reduce noise and displayed in a "copy number karyotype" with a choice of variability measures. A typical number of chromosomes used is 4. Significant deviations from the ratio value of 1.0 are noted as amplifications or deletions.

This work was funded by the US DOE contract DEAC0376SF00098.

[1] Piper J, Rutovitz D, Sudar D, Kallioniemi A, Kallioniemi O-P, Waldman F, Gray JW, Pinkel D. Computer Image Analysis of Comparative Genomic Hybridization. Cytometry 16 (in press)
[2] Ji L. Intelligent Splitting in the Chromosome Domain. Pattern Recognition 22: 519-532, 1989