Minisymposium 21: Emerging technologies
Abs #
42005: Computer-assisted morphometric analysis of Arabidopsis development
We continue to develop a computer-aided morphometric tool that measures small changes in Arabidopsis seedling form with minute resolution. It is based upon the automated acquisition and analysis of high-resolution infrared CCD images, providing a non-invasive means for determining the kinetics of visible developmental processes (Curr Opin Plant Biol 4: 436). Earlier and current versions of this tool have been used to gauge the roles of various developmental photoreceptors and downstream signaling elements in light-controlled hypocotyl growth (PNAS 96: 14142; Plant J 26: 471; Plant J 28: 333; Plant Physiol 126: 1291). For previous versions, however, automated growth analysis was limited to the first 2h following light treatment because they tracked growth simply as the difference in seedling height between successive images. Even though this method could faithfully assess growth over initial periods, other light-induced morphological changes occurring later, such as hook opening, eventually interfered with accurate measurements of length changes. Recent advances now enable fully-automated determinations of seedling length through time, regardless of these other developmental changes. Briefly, images processed to high contrast are analyzed through a multistep computer algorithm that identifies and measures seedling hypocotyl midline length as follows. 1) Seedling edge is defined as a subset of coordinates stored as an unordered contour set. 2) A set of circles is then found composed of those that lie within seedling image space and having at least two points (separated by at least 2π/3 radians) residing in the edge contour. 3) Circle set center points are stored in a midline subspace. 4) The midline set is re-numerated from the seedling base in the image to the tips of the cotyledons. 5) Length is calculated as a summation of the Euclidean metric applied over the midline set. Flexibility is also incorporated into the framework of this algorithm to permit morphometric analysis of other important seedling features that change during development. These include, but are not limited to root elongation and development, apical hook opening, phototropic curvature, and cotyledon expansion. We will provide a detailed overview of this algorithm and corresponding results demonstrating its accuracy. We will also address our plans to make this tool available to the scientific community interested in gene function as it pertains to plant development.
