New automatized approach for image processing in Xenopus laevis oocytes

OLIVER GIBSON; FLORENCIA GUASTAFERRI; AGUSTÍN CAVIGLIA; GABRIELA AMODEO; MARCELO OZU; LUCIANO GALIZIA

Congreso; Primeras Jornadas Virtuales de la Sociedad Argentina de Biofísica; 2020

Sociedad Argentina de Biofísica

The Xenopus laevis oocyte is a cellular model used widely for aquaporin and ion channelheterologous expression. This cellular model allows recording sequential images orvideos with low-cost videomicroscopy equipment. Our group developed differentacquisition devices to study various aspects of the aquaporin functionality. Although thismodel would potentially allow obtaining data from thousands of oocytes from each donorfrog, one bottleneck in the experimental workflow is image processing. Timelapsesrecorded at high temporal resolution (e.g., 30 fps), may contain thousands of imagesfrom which cell volume information must be extracted. The traditional processing ofthose images is performed manually, using free software (e.g., ImageJ) to segment theoocyte and to evaluate cell volume under high circularity criteria. Oocyte segmentationis user-dependent and requires training and experimental criteria. Then, informationabout the oocyte area in each image is converted to a relative volume curve. The waterosmotic permeability coefficient (Pf) is then calculated using the slope of the Vt/V0 curve.These sequential steps are time-consuming. The challenge of this work is to present anew Phyton-based tool which allows processing images, extract the information andcalculate Vt/V0 curves and the P f of each oocyte. We developed this new script toproduce an automatic, faster, and operator-independent analysis of Pf. The results arecompared with the manual/traditional approach.