Rapid analysis of gene copy numbers by FISH and automated spot counting from stacked confocal images
Conference Paper
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
Fluorescent in situ hybridization (FISH) is an excellent method for detection of gene copy number alterations in cancer and other diseases. A limitation of the technology is the tedious, inaccurate and often highly subjective spot counting. For a number of reasons, automation of FISH spot counting has not been accomplished. FISH signals are often at different focal planes, resulting in interfering out-of-focus light. This paper describes current progress towards automated FISH spot counting, with particular reference to the previous technical limitations. A confocal digital imaging system was used to acquire a series of up to twenty-four consecutive confocal fluorescent images (along the Z-axis) of the 4m thick tumor tissue sections. Dual-color FISH with a test probe and a reference probe labeled in two different colors was applied. A new algorithm was applied that identified the spots on each level along the Z-axis, and then differentiated spots located vertically above the X-Y plane. The ratio of test to-reference probe signals produced an estimate of the gene copy number alterations in the specimens.
name of conference
Molecular Imaging: Reporters, Dyes, Markers, and Instrumentation