Fig. 2: Automated detection and quantification of spectrin ring periodicity in neuronal axons.

a SReD-based analysis pipeline. Calculate neuronal segment angles: SReD is used to determine axon orientations by detecting repetitions of reference blocks containing lines at different orientations in ‘skeletonized’ axons. Optimise reference block: An optimised reference block resembling a periodic ring pattern block is generated by designing test blocks with different combinations of inter-ring spacing and ring height and using SReD to determine the degree of repetition of each test block in several test input images. Autocorrelation functions are calculated for each repetition map, and the first harmonics’ amplitudes are used as a measure of how well each test block fits the test data’s periodic patterns. The set of parameter values with the highest average amplitude is chosen. Map rotation-sensitive block repetition: Block repetition analysis is used to find repetitions of the optimized block at various orientations. Quantitative analysis: The repetition maps are analysed by calculating autocorrelation functions and determining the average inter-ring spacing and pattern prominence from the autocorrelations’ first harmonics’ positions and amplitudes. All repetition analyses performed using the Pearson’s correlation coefficient metric. b Control (CTRL) dataset image: STORM localization density (grey) overlaid with SReD repetition maps (magenta). Insets: (i)’Angles’ - axon skeletons colour-coded by orientation; (ii)’High order’ - repetition map calculated using a 9-ring reference block; (iii)’Low order’—repetition map calculated using a 3-ring reference block. Scale bar: 5 µm. c Repetition maps comparing CTRL and swinholide A-treated (SWIN) groups. SWIN-treated samples show reduced periodic structures. Scale bar: 1 µm. d Quantification of axon segments with ring patterns. Box plots show a significant reduction in pattern-containing segments in SWIN vs. CTRL (n = 6 per group, mean ± SEM; CTRL: 0.694 ± 0.008, SWIN: 0.421 ± 0.007; p < 0.001, two-sided unpaired t-test). Box plot elements are represented as such: center line as the median (50th percentile); box limits as upper (75th) and lower (25th) quartiles; whiskers indicate minimum and maximum values. Source data are provided as a Source Data file.