Detection and tracking of clathrin-coated pits in biological images

Abstract

Dynamically tracking hundreds of individual pits is essential to determine whether there exist "hot spots" for the formation of clathrin-coated pits or if the pits formed randomly on the plasma membrane. We propose an automated approach to detect these particles based on an improved á trous wavelet transform decomposition with automatic threshold selection and post processing solution, and to track the dynamic process with a greedy algorithm. The results indicate that the detection method can successfully detect most particles in an image with accuracy of 98.61% and 97.65% for adaptor and clathrin images, respectively, and that the tracking algorithm can resolve merging and splitting issues encountered when analyzing dynamic, live-cell images of clathrin assemblies. Recently, much attention has been focused on the development of segmentation and particle tracking techniques for the analysis of large volumes of microscopy fluorescence images acquired during cellular level cell imaging [1,2]. Particle tracking is one method of acquiring information about cellular dynamics. Clathrin-coated pits can be found in all nucleated cells, and provide an important means by which proteins and lipids are removed from the plasma membrane (endocytosis) and transported to an internal compartment (endosome) [3]. Using electronic microscopes, fluorescent-labeled versions of a variety of marker proteins have provided a tantalizing glimpse into the dynamics of the system in living cells. The clathrin pathway has thus acquired special status for analyzing molecular mechanisms in membrane traffic. A single image may capture hundreds or thousands of coated pits forming at the surface of a cell, all of which may not behave in the same way in time and space. Thus, information about these objects may need to be extracted and assembled at different time points and correlated with information from other images obtained at different spectral frequencies. For that purpose, clathrin light chains can be tagged with fluorescent proteins such as enhanced green fluorescent protein (EGFP) or yellow fluorescent protein (YFP) [1], and long time-series data can then be acquired using wide-field, confocal and total internal reflection fluorescence (TIRF) microscopy. Owing to the large quantity of images and hundreds of coated pits in each obtained image, an automated analysis approach is therefore necessary to visualize such molecular imaging data.The process for computerized analysis of cellular microscopy images generally consists of auto-segmentation, tracking and feature extraction. The detection of particles is the most critical step in molecular or sub-cellular image analysis where knowledge of the morphology of particles and the distribution of fluorescence signals in the particles is required. Although several particle detection methods based on segmentation have been developed [4,5], results have varied. Under-segmentation and/or over-segmentation can occur because these methods are based only on the intensity of the images or the morphologica...