Fig. 4

Quantitative evaluation of mitochondrial morphology revealed a striking difference between wild-type and mmi1Δ mutant cells after oxidative stress. Mitochondria as shown in Fig. 2 and segmented by the MitoS_yeast deep learning model were quantitatively analyzed by ImageJ/Fiji analyses performed by macro 2. The results of output parameters: a) mitochondrial area to the cell area, b) fragment length, and c) circularity are presented as box plots with Tukey whiskers, means (+), medians (lines), and circles as outliers. The y-axis for fragment length is on the log2 scale. d) The results of filamentous factor are presented as the means ± SEs. The nonparametric Kruskal‒Wallis test, followed by Dunn´s multiple comparison test, was used to test the differences among the samples. The data of more than 300 cells from two independent experiments were analyzed for each treatment. e) Comparison of the MitoA analysis with the ImageJ/Fiji analysis run by macro 3. The parameters such as perimeter, major axis length, minor axis length, and solidity were tested. The results are presented as the means per figure (n = 8) and are organized in box plots with maximum and minimum whiskers, all calculated values (circles), means (+), and medians (lines). One-way ANOVA was used to test the difference between the samples. The correlation analysis of the results of the MitoA analysis and the ImageJ/Fiji analysis (macro 3) is shown with the calculated Pearson´s correlation coefficient (r). ***** P ≤ 0.0001, *** P ≤ 0.001, ** P ≤ 0.01, * P ≤ 0.05, ns, not significant.