Analysis code for: Curcumin Combined with Swimming Training Ameliorates Experimental Osteoarthritis in Rats: An Untargeted Serum Metabolomics Study

This dataset is the R script used for the phenotypic analyses reported in the article "Curcumin Combined with Swimming Training Ameliorates Experimental Osteoarthritis in Rats: An Untargeted Serum Metabolomics Study".The script reproduces the OARSI cartilage score analysis (Kruskal-Wallis test and Cliff's delta, Table 1), the subchondral bone micro-CT analysis (one-way ANOVA with LSD post-hoc tests and Holm adjustment of the seven prespecified contrasts, Table 2), the cartilage cytokine analysis (one-way ANOVA with LSD for IL-1beta; Welch's ANOVA with the Games-Howell procedure for TNF-alpha, whose variances were unequal, Table 3), the 2x2 curcumin x swimming factorial analysis with an exact permutation test of the interaction (Table 4), the complete pairwise contrasts of Supplementary Tables S4a, S4b, S4d, S4e and S4f, and Figure 6 and Supplementary Figures S1 and S2.The permutation test is exact rather than Monte-Carlo: all C(4n, 2n) allocations of the additive-model residuals are enumerated completely (924 allocations for n = 3 per cell; 2,704,156 for n = 6).

The script contains no random-number generation, so its output is identical on every run and on every platform.The script is self-contained: every individual animal value used in these analyses is embedded in the file, so no external data file is required. It depends only on base R, requires no additional package, has been tested under R 4.2.3, and runs in under one minute.Usage: Rscript OA_curcumin_swimming_analysis.RResults are printed to the console and written to ./results/Not covered by this script: the weekly body-weight analysis (repeated-measures ANOVA, performed in SPSS 26.0) and the untargeted metabolomics (XCMS v4.8.0 and R with ropls v1.30.0).