Microarray Imaging

For imaging DNA or protein microarrays in fluorescence, filters are added to the detection arm to block the reflected laser light, as well as fluorescence from other fluorophores on the specimen. For example, many microarrays use both Cy3 and Cy5. Cy3 is excited by a green laser (usually 532 nm), and Cy5 is excited with a red laser (often 635 nm). Although the red laser excites only Cy5, the green laser weakly excites Cy5 as well as strongly exciting Cy3. In addition, the emission spectra of Cy3 and Cy5 overlap slightly, so the reader manufacturer must choose filters carefully so there is no cross talk between detection channels. There may also be residual fluorescence from the substrate, which is minimized by confocal detection. Microarray scanners are available with up to four different laser sources, and multiple detection arms. Filter wheels containing a large number of filters allow sequential or simultaneous detection of several different fluorophores.

Several steps are required to image a microarray. First, the fluorophores used on the array are selected by the operator to enable the instrument to choose lasers and filter sets. A preview scan of the entire glass slide is performed for each fluorophore chosen, which allows the operator to choose the area to be imaged, and to set the detector gain and offset. The offset is set to minimize the background noise in the area of the image between the probe spots—this should be set to a small value, but not to zero. The gain is then usually set so that the brightest probe spots in the image are just below detector saturation.

For some applications, the intensity of images from different fluorophores is adjusted by changing laser intensities. Finally, a high-resolution scan is performed on the area of the slide containing the microarray probe spots. For common microarrays with probe spots of 100 mm or more in diameter, a pixel setting of 5 mm works well. For best results, the pixel size should match the diameter of the focused laser beam. In some scanners, the smallest pixel size available is smaller than the focal spot diameter, so choosing this setting will not improve the image resolution, but it does act like frame averaging by making a larger number of overlapping intensity measurements inside the area of each probe spot.

The objective of these measurements is to analyze the fluorescence images to provide the user with a table of numbers (for each fluorophore) representing the integrated fluorescence intensity of all of the probe spots on the microarray. These numbers can then be transferred to an analysis program that matches the user's specific requirements. Image analysis comprises several steps. First, measurement circles in a grid pattern matching the microarray grid are placed around every probe spot in the image. Because the probe spots are not always centered exactly on the grid positions, the position of the measurement circles is adjusted to match the actual probe spot positions. The fluorescence intensity for each probe spot is calculated by integrating the fluorescence intensity of the pixels inside each measurement circle, and then subtracting the local background intensity calculated from pixels in the area just outside each circle. One number, representing the integrated fluorescence intensity for each probe spot and for each fluorophore, is stored in a table which is then available for the user's specific application software.

A trained operator can perform all of the steps described above to image a microarray like that shown in Fig. 4 in both Cy3 and Cy5, and analyze the images to produce tables of integrated intensities in an hour. A fully automated reader performs the same task in 10 min, and the only interaction with the operator is to load the microarray and specify which fluorophores are used. Ten minutes later, the operator can retrieve the output data files.

Figure 4A shows Cy3 and Cy5 images of a DNA microarray on a glass microscope slide, imaged with 5-mm pixels (and 5-mm focal spot size) on a DNAscopeTM IV. The scan area was 20 x 70 mm. Figure 4B (left) shows a magnified image of one subgrid from the image in Fig. 4A, and Fig. 4B (right) illustrates the placement of measurement circles by the automated analysis routines in MACROviewTM. Note that the automated detection and placement algorithm has correctly ignored the smear of fluorescence at the edge of the picture and the dust particles in the image. Many grid placement algorithms require a bright fluorescent probe spot at the top left of

Was this article helpful?

0 0
Getting Started With Dumbbells

Getting Started With Dumbbells

The use of dumbbells gives you a much more comprehensive strengthening effect because the workout engages your stabilizer muscles, in addition to the muscle you may be pin-pointing. Without all of the belts and artificial stabilizers of a machine, you also engage your core muscles, which are your body's natural stabilizers.

Get My Free Ebook


Post a comment