![]() Working in vivo or in situ with Drosophila is one of the main reasons behind using it as a model organism. Drosophila is a powerful model organism generally used to investigate gene function, developmental processes and model human diseases. Additionally, results are reproducible: automatic programs perform consistently and always yield the same cell count for a given sample regardless of the number of times it is counted. Quantification is automatic, accurate, objective and fast, enabling reliable comparisons of multiple specimens of diverse genotypes. We have developed a range of publicly available methods that can count the number of dividing or dying cells, neurons or glia, in intact specimens of fruit-fly Drosophila embryos ( Forero et al, 2009, 2010, 2010a). Cell Profiler ( Carpenter, 2006) enables combinations of image-processing methods that can be used to count cells, but it is not very user friendly for most biologists as it requires computation expertise. In any case, counting all nuclei is not always most informative, as it does not qualify on cell type (is the number of neurons or glia altered?) or cell state (do the changes affect dividing or dying cells?). Identifying all the nuclei is extremely challenging from the point of view of imaging because cells can be tightly packed. Some automatic techniques have been developed to segment cell nuclei from mammalian tissue sections or from whole Drosophila brains in 2D and 3D images ( Lin et al., 2003 Shimada et al., 2005 Wählby 2003 Wählby et al., 2004), but they are not useful to analyse large sample sizes because the intensive computation slows down the process. Such methods are challenging, as they require large stacks of images to capture the whole sample, and can encounter greater difficulty in distinguishing labelled cells from background signal. Whereas excellent automated methods can be purchased commercially and are widely used to count cells after dissociation or in cell culture, fewer methods have been developed to count cells in situ or in vivo. The advent of confocal microscopy, which allows the capture of 3D images, has enabled the development of automatic and semi-automatic image processing methods to count cells in whole tissues or entire small animals. Manual counting can be experimentally cumbersome, tedious, labour intensive and error prone. These methods can be extremely time-consuming, estimates can be inaccurate, and the questions that can be addressed using these methods are limited. Counting in vivo or in situ is generally carried out manually, or it consists of estimates of number of cells stained with a particular cell marker or inferences from anatomical alterations. in the intact animal) or at least in an entire organ or tissue (i.e. ![]() To maintain information relevant to how genes and cells behave in the organism, it is best to count cells in vivo (i.e. However, these approaches alter the normal cellular contexts and the procedures themselves can alter the relative numbers of cells. fluorescence-activated cell sorting, FACS, based), or when they are distributed in a dish in cell culture experiments, using image processing techniques in 2D (e.g. Generally, cells are counted using automated methods after dissociating cells from a tissue (e.g. Thus to understand normal animal development, injury responses and disease, it is important to find out how many cells die or divide, or how many cells of a given type there are in an organ. spinal cord injury) results in an increase in cell death, plus a homeostatic regulation of cell proliferation. Cell number is the balance between cell division and cell death it is controlled tightly during growth and it can be altered in disease, most notoriously neurodegeneration and cancer. One useful parameter to quantify is cell number. Image-processing methods have enormous potential to extract information from this kind of samples, but surprisingly, they are still relatively underexploited. For this, they look at how alterations in gene function and application of drugs affect tissue, organ or whole body integrity, using confocal microscopy images of samples stained with cell specific markers. Biologists aim to understand how cells behave and what genes do to build a normal animal, and what goes wrong in disease or upon injury. ![]() Image processing methods have opened the opportunity to extract quantitative information from confocal microscopy images of biological samples, dramatically increasing the range of questions that can be addressed experimentally in biology.
0 Comments
Leave a Reply. |