An optimized, fully automated system for fast and accurate identification of chromosomal rearrangements by multiplex-FISH (M-FISH)

TLDR

Multiplex‑FISH (M‑FISH) enables each of the 24 human chromosomes to be uniquely stained with distinct colors, and the addition of new fluorochromes expands its chromosomal resolution. The study aims to present several significant optimizations of the M‑FISH technology. The authors introduced a combinatorial labeling strategy, a fully motorized microscope with an eight‑position filter wheel, and a novel multichannel image‑analysis method that automatically determines spectral signatures and incorporates built‑in quality controls. These optimizations allow a complete metaphase spread analysis in under three minutes while the built‑in quality controls demonstrate the reliability of automated chromosome classification.

Abstract

Multiplex-FISH (M-FISH) is a recently developed technique by which each of the two dozen human chromosomes—the 22 autosomes and the X and Y sex chromosomes—can be stained or “painted” with uniquely distinctive colors. Using a combinatorial labeling technique and a specially designed filter set, each DNA probe can be identified by its unique spectral signature. Here we present several significant optimizations of the M-FISH technology. First, a new strategy for labeling the probes is described which allows for easy and fast production of the complex M-FISH probe mix. Second, a newly developed, completely motorized microscope equipped with an eight-position filter wheel and a new generation of filter sets is presented that allows fully automatic imaging of a complete metaphase spread within seconds. Third, to determine the characteristic spectral signatures for all different combinations of fluorochromes, we developed a novel multichannel image analysis method. The spectral analysis is solely guided by the image information itself and does not require any user interaction. A complete analysis of a metaphase spread can be accomplished in less than 3 min. Sophisticated built-in quality controls were developed, and the value of visual inspection of M-FISH images as a simple means of controlling the computer-generated chromosome classification are illustrated. In addition, we discuss advantages of adding new fluorochromes to the traditionally used five fluorochromes.