Primary result: distance from baseline after ciprofloxacin
Bray–Curtis distance between each mouse’s window samples (days 8–12) and its own day-0 sample, averaged within mouse and then within cage. The analysis plan was frozen before any diet comparison.
−0.145Low-MAC − standard contrast (cage-equal)
2 of 2low-MAC cages below all three standard-diet cages
17 / 5eligible mice / cages
Overlapindividual mice overlap between diets
Figure 1
The cage pattern was stable across the tested Bray–Curtis versions but depended on the distance metric.
A: Bray–Curtis distance from each mouse’s own day-0 sample, mean of days 8–12. Points are mice, bars are cage means, dashed lines are the cage-equal diet means. B: the PD − STD contrast in every distance-from-baseline analysis, as % of the STD mean (the metrics have different scales); the contrast in the original units is printed on the right. Percentages express each contrast relative to its own STD mean; they do not represent percentage recovery. 218/166 = forward/reverse read truncation lengths (sensitivity setting; primary 200/150). PD = low-MAC diet (the paper’s MD arm); STD = standard diet.
Cage values
Diet
Cage
Mice
Cage mean
Standard
H_c1
4
0.897
Standard
H_c2
4
0.812
Standard
H_c3
2
0.882
Low-MAC
H_c4
4
0.712
Low-MAC
H_c5
3
0.724
Diet means: standard 0.864, low-MAC 0.718 (Δ = −0.145).
How to read this
A lower value means closer to the mouse’s own pre-antibiotic composition during the window; it is not a direct measure of recovery rate.
All standard-diet day-0 samples were sequenced in run 1. Low-MAC day-0 samples and all analysed days 8–12 samples were sequenced in run 2.
Baseline sequencing run is confounded with diet, cohort concurrency is unverified, and only five cages contribute. Results are therefore reported descriptively, and the pattern cannot be attributed to diet.