Photo Phenosizer, a rapid machine learning-based method to measure cell dimensions in fission yeast.

Abstract

Cell metrics such as area, length, and width provide informative data about cell cycle dynamics. Factors that affect these dimensions include environmental conditions and genotypic differences. Fission yeast ( Schizosaccharomyces pombe ) is a rod-shaped ascomycete fungus in which cell cycle progression is linked to changes in cell length. Microscopy work to obtain these metrics places considerable burdens on time and effort. We now report on Photo Phenosizer (PP), a machine learning-based methodology that measures cell dimensions in fission yeast. It does this in an unbiased, automated manner and streamlines workflow from image acquisition to statistical analysis. Using this new approach, we constructed an efficient and flexible pipeline for experiments involving different growth media (YES and EMM) and treatments (Untreated and MMS) as well as different genotypes ( cut6-621 versus wildtype). This methodology allows for the analysis of larger sample sizes and faster image processing relative to manual segmentation. Our findings suggest that researchers using PP can quickly and efficiently determine cell size differences under various conditions that highlight genetic or environmental disruptions.