Automatic phenotyping of multi-channel Schizosaccharomyces pombe images

Schizosaccharomyces pombe shares many genes and proteins with humans and is a good model for chromosome behavior and DNA dynamics, which can be analyzed by visualizing the behavior of fluorescently tagged proteins in vivo. However, performing a genome-wide screen for changes in such proteins requires developing methods that automate analysis of multiple images. We developed a high content analysis system to robustly segment transmitted illumination images, extract cell and nucleus boundaries, and quantitatively characterize the fluorescence within each compartment. A support vector machine (SVM) is trained to automatically judge if a cell is undergoing septation, and another two SVMs are trained to classify pombe cells into various phenotypes according to its cell shape and fluorescence signal profile. We applied this system to automatically calculate the percentages of cells of different phenotypes for 4000 S. pombe mutants.