Multifluor-fluorescence in-situ hybridization (M-FISH)...

Radiant energy – Luminophor irradiation

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C250S339020, C250S461200

Reexamination Certificate

active

06225636

ABSTRACT:

BACKGROUND OF THE INVENTION
The present invention relates generally to fluorescence microscopy, and more specifically to improvements in image alignment where multiple multiband filters are used in M-FISH techniques.
M-FISH (which stands for multifluor-FISH, multiplex-FISH, multicolor-FISH, or multispectral-FISH) is a technique by which a number of fluorochromes (sometimes referred to as fluorescent dyes, or simply dyes) are used in what is otherwise a standard FISH (Fluorescence In-Situ Hybridization) procedure. A FISH sample is prepared by using multiple probes, each of which binds to a different DNA sequence in the chromosomes in the sample. Each probe is labeled with a different dye or combination of two or more dyes.
A given dye is characterized by an excitation (absorption) spectrum and an emission spectrum. The excitation and emission spectra are also sometimes referred to as the excitation and emission bands. Accordingly when the dye is irradiated with light at a wavelength within the excitation band, the dye fluoresces, emitting light at wavelengths in the emission band. Thus when the sample is irradiated with excitation radiation in a frequency band that excites a given dye, portions of the sample to which the probe labeled with the given dye is attached fluoresce. If the light emanating from the sample is filtered to reject light outside the given dye's emission band, and then imaged, the image nominally shows only those portions of the sample that bind the probe labeled with the given dye.
FIG. 1
is a schematic showing a representative epi-illuminated fluorescence microscope system 5 for generating a single FISH image of a sample 10. The optical arrangement in system 5 includes an excitation filter 12 (shown as one of several such filters on a filter wheel), a dichroic mirror 15, a microscope objective 17 (say 60-10OX), and an emission filter 20 (sometimes also referred to as a barrier filter). Excitation radiation from a source 25 passes through excitation filter 12, is largely reflected by dichroic mirror 15 and proceeds through the microscope objective 17 to sample 10. The excitation light traveling toward the sample is shown schematically by hollow arrowheads. Fluorescent radiation emitted from sample 10 passes back through objective 17, through dichroic mirror 15, through emission filter 20 to form an image in an image plane 30. The fluorescent light traveling away from the sample is shown schematically by solid black arrowheads. The image is digitized by a CCD camera 32 and the digitized image is sent to a computer 35 for subsequent processing.
If the filters have single passbands, the particular filters and dichroic mirror are specific to a single dye in the sample. Images for other dyes in the sample are acquired by substituting optical elements configured for the excitation and emission bands for each other dye. The dichroic mirror and the emission filter are typically rigidly mounted to a supporting structure 40 (shown in phantom), often referred to as a cube, with multiple cubes being movable into and out of the optical path. oppositely directed arrows 42 represent a suitable mechanism such as a rotatable turret or a detented slide mechanism. The multiple excitation filters are typically deployed on a rotatable filter wheel (as shown).
The fact that the different images are produced by moving different cubes into the image path inevitably causes lateral and focus shifts and the like, thereby leading to misregistration of the images. While there are well-known image registration techniques, such techniques tend not to be robust. Since the different images represent fluorescence from different dyes, whose probes are attached to different portions of objects in the sample, attempts to register the images are prone to failure under at least some circumstances.
A number of variations of the instrumentation are well established in the prior art. For example, [Castleman93] discloses the use of a color CCD camera, multiband excitation and emission filters, and a polychroic mirror for simultaneously digitizing emissions from specimens labeled with three fluorescent dyes. The excitation filter passes three narrow bands corresponding to the excitation bands of the three dyes, the polychroic mirror reflects the three excitation bands while transmitting the corresponding emission bands, and the emission filter only passes wavelengths falling within the three emission bands.
As described in [Castleman93], there is significant cross-talk between the fluorescence channels due to the inevitable overlap among dyes' emission spectra and the camera's sensitivity spectra. This is addressed by an image processing step, referred to as color compensation, based on a predetermined knowledge of how each of a given dye's emission is recorded in each of the camera's RGB channels. A 3×3 matrix is determined, and the inverse of this matrix is applied to measured RGB values to eliminate the effect of color spread among the camera RGB channels.
Using a color camera and the multiband filters provides three images with the same cube, and thus avoids the registration problem. On the other hand, color cameras normally provide lower resolution and lower sensitivity than monochrome cameras. [Bornfleth96] compares the traditional technique of acquiring three monochrome images and the technique using a color camera and multiband filters, and concludes that the results are comparable.
The use of a single cube with a color camera is limited to three dyes, however, whereas it is often desired to have more than three dyes. For example, five to seven dyes could provide a significantly larger number of possible combinations. While it may be possible to design a single cube configured for this many dyes, this would create additional challenges to recognize uniquely which dye or combination of dyes contributes to each camera channel. The already stringent optical demands would become even more stringent. It is necessary to have narrowband emission filters since the polychroic mirror typically transmits a small percentage (say 5% to 10%) of the much stronger excitation radiation that is inevitably reflected and scattered from the microscope surfaces and the sample.
While it would be possible to extend the multiband approach of [Castleman93] to more than three dyes by providing two or more multiband cubes and multiband excitation filters, each configured for three dyes, the problem of image registration would again arise.
SUMMARY OF THE INVENTION
The present invention provides techniques for acquiring and analyzing M-FISH images in a manner that allows for proper image registration and yet is easily scaleable for large numbers of dyes. The invention has applicability with epi-illumination and trans-illumination optical systems. The invention further provides techniques for combining a plurality of registered images to provide labeled images.
In short, with respect to epi-illumination embodiments, an aspect of the invention recognizes that it is impractical to configure a multiband filter cube for more than four or five dyes, and therefore takes the approach of having multiple filter cubes for different subsets of the entire set of desired dyes. The invention addresses the registration problem by configuring the different multiband cubes so that pairs of cubes have a common dye for which they are configured. Accordingly, a first set of images generated with a first cube includes an image that corresponds to (contains the same features as) one of the images in a second set of images generated with a second cube. This allows the first set of images to be precisely and accurately aligned with the second set of images.
In a specific embodiment, there are two cubes, each configured for four dyes with one of the dyes being the same for the two cubes. Further, it is preferred to have the dye that is common to the two cubes to be a counterstain that stains all DNA and therefore provides an image with substantially all the objects, thereby a

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Multifluor-fluorescence in-situ hybridization (M-FISH)... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Multifluor-fluorescence in-situ hybridization (M-FISH)..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multifluor-fluorescence in-situ hybridization (M-FISH)... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2437621

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.