Extended Data Fig. 9: Quantitative analysis of whole cell 3D super-resolved mitochondria images.
From: Field-dependent deep learning enables high-throughput whole-cell 3D super-resolution imaging
a, Overview of 16 full frame images of whole cell 3D super-resolved mitochondria. b, Magnified angled views of 3 representative cells denoted by the boxes in a. c, Quantitative analysis of mitochondria morphology and network connectivity using Mitochondria Analyzer36. n = 121 cells imaged from 16 independent experiments and then classified by k-means++ into 3 groups. The numbers of cells for type 1/2/3 are 47/22/52, respectively. One-way ANOVA with two-sided Bonferroni multiple comparison test, the significance level was set as 0.05. The adjusted P-value for pairwise comparison is shown in each sub-panel of c. Boxplots are drawn from the 25th to 75th percentile with the horizontal line denoted the median and the whiskers extending to the minimum and maximum. Values outside 1.5 times interquartile range are defined as outliers which are shown as magenta circle. d, Principal component analysis was performed on the features used in c with the first 3 principal components plotted, color indicates the categorization results. Scale bars, 50 μm (a) and 10 μm (b1).
