R 4.5.1 | limma 3.66.0 | edgeR 4.8.2 | cores 6 [setup] voom has argument adaptive.span: TRUE | its default TRUE ==== 1. Deposited files ==== [data] four deposit files in data/, MD5 checked against the OSF checksums: ok [data] count matrix rows (genes) x columns (libraries): 13230, 27 [data] all cells non-negative integers: TRUE | NA cells 0 | identical library columns 0 [data] sample names identical and in the same order: counts vs sample table, counts vs normalised matrix: TRUE, TRUE [data] duplicated row names: 0 | rows sharing a gene symbol with an earlier row 1 [data] fitness group x sex: sex Fit male_0 female_1 Fit_0 3 9 Unfit_1 13 2 [data] spike-in rows (RNA_SPIKE_): 76 | of which detected in >= 1 sample 75 | all-zero rows 946 | rows detected in <= 5 samples (%) 26.8 [data] PHV offset mean (SD): fit: -1.76, 0.8 | unfit -2.47, 0.96 | all -2.15, 0.94 | range -4.18, -0.59 ==== 2. The authors' script, run as deposited ==== [A2] script, current limma default: sum(adj.P.Val < 0.05): 273 | coefficient < 0: 147 | > 0: 126 | nominal p < 0.05: 1079 [A2] max |script - own pipeline|: logFC, adj.P.Val: 0 0 [A2] max |voom E from script - deposited normalised matrix|: 4.97e-14 [A2] script, voom span 0.5 fixed (pre-3.66 default): sum(adj.P.Val < 0.05): 257 | coefficient < 0: 145 | > 0: 112 | nominal p < 0.05: 1092 [A2] max |script - own pipeline|: logFC, adj.P.Val: 0 0 [A2] max |voom E from script - deposited normalised matrix|: 4.97e-14 [A2] script mentions calcNormFactors / TMM: FALSE [A2] limma NEWS: 3.66.0 The default for `adaptive.span` in voom() and [A2] hand-coded pre-3.60 voom vs voom(span=0.5, adaptive.span=FALSE): max |E diff|, max |weight diff|: 0 0 [A2] span chosen by the current default (adaptive): 0.409 ==== 3. Printed numbers against specifications ==== Printed (author manuscript): 256 genes, 145 up and 111 down in fit, log2FC -5.83 to 6.55, 33 top genes in Figure 2 (22 up, 11 down), 108 arrow entries for 57 genes in Tables 2-3; Supplementary Table 1 lists 256 genes. Coefficient < 0 is printed as 'up in fit'. [A3] Tables 2-3 entries: 108 | unique genes 57 [A3] Table S1 genes: 256 [A3] Figure 2 labels up/down: 22, 11 spec hits neg pos min max n15 tab108 arrow108 fig2 spike S1_of_256 S1_dlfc S1_rel_fdr A current default 273 147 126 -5.82 6.55 35 108 108 FALSE 2 255 7.6e-02 2.3e+00 B span 0.5 fixed 257 145 112 -5.83 6.55 34 108 108 FALSE 1 256 4.9e-09 2.9e-03 C = B minus spike row 256 145 111 -5.83 6.55 33 108 108 TRUE 0 256 4.9e-09 2.9e-03 D current + TMM 256 145 111 -5.82 6.55 35 107 108 FALSE 1 252 9.4e-02 1.8e+00 E span 0.5 + TMM 258 145 113 -5.83 6.56 34 107 108 FALSE 1 253 3.3e-02 1.4e+00 F span 0.5, robust eBayes 256 145 111 -5.83 6.55 33 107 108 FALSE 1 255 4.9e-09 2.4e+04 Figure 2 set (|log2FC| >= 1.5 among hits) against the 33 labelled genes: spec fig2_extra fig2_missing A current default FASN,H1FX,RNA_SPIKE_ERCC-00143 OTUB2 B span 0.5 fixed RNA_SPIKE_ERCC-00143 C = B minus spike row D current + TMM CCDC71L,CHFR,H1FX,RNA_SPIKE_ERCC-00143 OTUB2,MTX3 E span 0.5 + TMM H1FX,RNA_SPIKE_ERCC-00143 MTX3 F span 0.5, robust eBayes RNA_SPIKE_ERCC-00143 MAL S1_of_256 = Table S1 genes that are hits in the specification; S1_dlfc = largest |S1 log2FC + coefficient|; S1_rel_fdr = largest relative gap between S1 FDR (3 digits) and adj.P.Val. [A3] spike-in row among the span-0.5 hits: RNA_SPIKE_ERCC-00143 | coefficient 4.94 (positive = printed 'down in fit', so removing it leaves 145 / 111) [A3] set of C equals the 256 genes of Table S1: TRUE [A3] TOX2 (Table 3, hypertension row) adj p: A, B, D, E: 0.0419, 0.0416, 0.1270, 0.0885 | reads 3 | samples 2 [A3] table entries not significant: D: TOX2 | E TOX2 | F MAL [A3] max |voom E - deposited matrix|: A (no TMM), D (TMM): 4.97e-14 2.18e-09 (quantile normalisation removes most of TMM from E) [A3] TMM factors, range: 0.843, 1.188 Hit counts by variant of the shipped analysis (3-way coefficient): all_rows spikes_removed_before_fit zero_rows_removed robust_eBayes trend_eBayes no_quantile TMM current_default 273 271 74 272 276 1 256 span_0.5_fixed 257 258 78 256 272 1 258 ==== C1. Tested coefficient and the additive model ==== [C1] design columns: 8 | rank 8 | residual df 19 [C1] tested column (script: coef = ncol(design)): FitUnfit_1:sexfemale_1:PHV [C1] script lines using coef = ncol(design): 2 [C1] SCO2: 3-way coefficient equals (unfit F - unfit M) - (fit F - fit M) PHV slopes: -3.26524, -3.26524 [C1] additive model ~ Fit + sex + PHV, current default: genes 13230 | hits 0 | min adj p 0.407 | raw p < 0.05: 501 (3.8%) | -log10(min adj) 0.39 [C1] additive model ~ Fit + sex + PHV, span 0.5 fixed: genes 13230 | hits 0 | min adj p 0.53 | raw p < 0.05: 527 (4%) | -log10(min adj) 0.28 [C1] shipped coefficient: -log10(min adj p), current / span 0.5: 14.57, 14.5 [C1] arrows of Tables 2-3 equal minus the sign of the 3-way coefficient (of 108), current / span 0.5: 108, 108 [C1] table entries significant under the shipped coefficient (of 108), current / span 0.5: 108, 108 [C1] Fit main effect of the full model (boys at PHV = 0): hits: 162 | PHV = 0 outside observed range TRUE [C1] unique table genes significant under it (of 57): 23 | of those with sign opposite to the arrow 23 [C1] 256-list genes detected in >= 6 samples: 27 | printed direction equals raw fit-minus-unfit (voom log-CPM) sign in 7 | equals additive adjusted sign in 11 gene printed coef_3way adj_3way additive_fit_minus_unfit additive_p reads samples IL2RA down 5.83 0.015 -1.13 0.173 231 24 SCO2 up -3.29 0.026 0.32 0.437 9970 27 GRB2 up -2.33 0.021 0.22 0.446 874 27 MAL down 1.77 0.049 0.05 0.832 5556 27 GNG10 up -1.74 0.035 -0.42 0.073 29331 27 CREB3L3 up -1.54 0.000 0.30 0.102 4 1 PPP2R5E up -2.91 0.011 0.22 0.553 10 3 ==== C2. Where the hits are; leverage; filtering ==== [C2] leverage of the 2 unfit girls: 1, 1 | largest among the other 25 0.936 [C2] share of squared weight on the 3-way coefficient: 3 fit boys, 2 unfit girls: 0.49, 0.33 [C2] current: hits: 273 | detected in <= 5 samples 244 | median total reads 3 | reads only in the 2-child cell 108 | residual SD < 1e-6 108 | single-sample hits (unfit girls, fit boys, other) 105, 93, 0 [C2] span 0.5: hits: 257 | detected in <= 5 samples 229 | median total reads 3 | reads only in the 2-child cell 108 | residual SD < 1e-6 108 | single-sample hits (unfit girls, fit boys, other) 105, 80, 0 [C2] rows with reads only in the 2-child cell: 109 | of those in one child 105 [C2] filterByExpr (additive design) keeps: 6206 | shipped coef hits: current, span 0.5 0, 0 | min adj 0.231, 0.226 | additive hits 0, 0 | min adj 0.232, 0.233 [C2] span 0.5, detected in >= 2 samples: rows: 11331 hits 0 | zero rows removed: rows 12284 hits 78 [C2] eBayes prior df / variance: all rows: 0.65, 0.046 | zero rows removed 1.66, 0.406 [C2] span 0.5, no quantile normalisation: hits: 1 | distinct values in zero-count cells: with / without quantile 1, 27 [C2] spike-in rows with nominal p < 0.05 (of 76): shipped coef, additive (current): 9, 0 | span 0.5 9, 0 [C2] genes named in the Results' enrichment sentence: reads: 4, 10 | samples detected 1, 3 ==== C2a. Permutations (voom refitted every time; counts of genes at BH < 0.05) ==== Observed: 273 (current default), 257 (span 0.5). Schemes break real sex and maturity structure, so they approximate a null only. permutation fits: 1125 [C2a] permutation wall time (minutes): 10.1 [C2a] P1 within-sex expression, 30, current: median 196.0, IQR 181-220, range 139-371 | >= observed (273): 3 of 30, P 0.100 [C2a] P2 all columns, 200, current: median 204.0, IQR 175-233, range 125-443 | >= observed (273): 19 of 200, P 0.095 [C2a] P3 joint covariate rows, 40, current: median 208.0, IQR 172-240, range 108-368 | >= observed (273): 2 of 40, P 0.050 [C2a] P4 within-sex Fit labels, 200, current: median 185.0, IQR 162-222, range 102-430 | >= observed (273): 9 of 200, P 0.045 [C2a] P4 within-sex Fit labels, 200, span 0.5: median 174.5, IQR 154-205, range 93-356 | >= observed (257): 8 of 200, P 0.040 [C2a] P5 within-sex expression, 200, current: median 204.5, IQR 170-236, range 92-532 | >= observed (273): 27 of 200, P 0.135 [C2a] P5 within-sex expression, 200, span 0.5: median 190.0, IQR 159-215, range 93-495 | >= observed (257): 17 of 200, P 0.085 [C2a] P6 all 55 choices of the unfit-girl pair, span 0.5 : median 200 range 110 - 257 | observed pair 257 rank 1 of 55 [C2a] P1: additive model on the same 30 permuted data sets, runs with >= 1 hit: 11 of 30 | median 0 [C2a] P1: median: 196 | runs reaching the observed 273 3 ==== C3. 273 against 256 ==== [C3] current default: hits 273 | coefficient < 0 / > 0 147, 126 | log2FC range (printed sign) -5.82, 6.55 | |log2FC| >= 1.5: 35 | |log2FC| >= log2(1.5): 192 [C3] span 0.5 fixed: hits 257 | coefficient < 0 / > 0 145, 112 | log2FC range (printed sign) -5.83, 6.55 | |log2FC| >= 1.5: 34 | |log2FC| >= log2(1.5): 189 [C3] span 0.5, spike row removed from the list: 256 | coefficient < 0 / > 0 145, 111 | range -5.83, 6.55 | |log2FC| >= 1.5: 33 [C3] current + TMM: 256 145 neg / 111 pos | range -5.82, 6.55 [C3] genes shared: current vs current + TMM; span 0.5 minus spike vs current; vs current + TMM: 255, 255, 252 [C3] span 0.5 minus spike vs Table S1: genes in both, only here, only in S1: 256, 0, 0 [C3] current + TMM vs Table S1: genes in both, only here, only in S1: 252, 4, 4 [C3] current default vs Table S1: genes in both, only here, only in S1: 255, 18, 1 [C3] distinct log2FC values among the 256 in Table S1: 87 | distinct FDR values 78 ==== C4-C6. Table 1 ==== [C4] printed: PHV offset fit -1.76 (0.80) n = 12, unfit -2.47 (0.96) n = 15, p = 0.50 [C4] from printed statistics: Student, Welch: 0.051, 0.046 [C4] from deposited per-child PHV: Student, Welch, Mann-Whitney: 0.051, 0.046, 0.071 [C5] IL-6 (n = 25; group n 10-12 / 13-15; rounding corners): total mean: 2.02, 2.07 | total SD 1.05, 1.11 | Student p 0.274, 0.332 | Welch p 0.246, 0.290 [C5] printed total 2.31 (SD 1.98), p 0.26; printed total mean above both group means (1.79, 2.25): TRUE | printed SD inside implied range FALSE [C5] sentences with IL-6 and 'lower' in the article text: 2 [C6] printed: BMI group counts 3 of 12 fit, 3 of 15 unfit; p = 0.56 [C6] Pearson, Yates, Fisher: 0.756, 1, 1 [C6] Table 1 BMI-group row labels in the author manuscript: Overweight, Obesity (the audit describes one combined label in the journal version) [C6] Methods: 'EASE score < 0.05' present: TRUE | 'background' present FALSE | any adjustment mentioned in the EASE sentences FALSE ==== C7. SCO2, SPDYC, IL2RA ==== [C7] current default SPDYC 6.55 | SCO2 3.27 | IL2RA -5.82 | SCO2 rank among up (of 273 hits) 2, by |log2FC| 7 | IL2RA from bottom 1 | genes above SCO2 on the up side: SPDYC [C7] fold change >= 1.5 (|log2FC| >= log2 1.5): 192 of 273 | |log2FC| >= 1.5: 35 [C7] span 0.5, no spike SPDYC 6.55 | SCO2 3.29 | IL2RA -5.83 | SCO2 rank among up (of 256 hits) 2, by |log2FC| 6 | IL2RA from bottom 1 | genes above SCO2 on the up side: SPDYC [C7] fold change >= 1.5 (|log2FC| >= log2 1.5): 188 of 256 | |log2FC| >= 1.5: 33 [C7] current + TMM SPDYC 6.55 | SCO2 3.26 | IL2RA -5.82 | SCO2 rank among up (of 256 hits) 3, by |log2FC| 8 | IL2RA from bottom 1 | genes above SCO2 on the up side: CHFR,SPDYC [C7] fold change >= 1.5 (|log2FC| >= log2 1.5): 188 of 256 | |log2FC| >= 1.5: 35 [C7] among the 256 genes detected in all 27 samples (n): 23 | highest SCO2 3.29 | lowest PIK3IP1 -3.88 [C7] reads / samples detected: SPDYC, SCO2, IL2RA: 167, 9970, 231, 18, 27, 24 [C7] Figure 2 caption says 'fold change' / mentions log2: TRUE, FALSE | Table 3 note mentions log2FC TRUE | Results mention log2FC ranged TRUE [C7] Discussion names SCO2 and IL2RA as highest and lowest: TRUE ==== C8. Methods and what the analysis chose ==== [C8] occurrences in the author manuscript text: voom 0 | interaction 0 | spike 0 | ERCC 0 | filter 0 | TMM 0 | edgeR 0 | calcNormFactors 0 | background 0 | limma version 0 | R version 0 | sessionInfo 0 [C8] occurrences of 'version': 3 [C8] rows tested: 13230 | spike-in rows 76 | all-zero rows 946 | rows after removing spike-ins and all-zero rows 12209 ==== C9-C10. The deposit ==== [C9] script reads: Fit_Unfit_BW.txt, coldata.txt | present in the deposit FALSE, FALSE [C9] script writes: Fit_Unfit_BW_sex_PHV_voom_quantiles.txt, voomQuantiles.txt, vQuantWeights.txt | present in the deposit FALSE, FALSE, FALSE [C9] deposit holds: coldata.txt.txt, Fit_Unfit_BW_raw.txt.txt, ScriptR_RNAseq_limma_FitUnfit.R, voomQuantiles_normalized.txt.txt [C9] library calls: library(edgeR), library(DESeq2) | sessionInfo in script FALSE | edgeR Depends R (>= 3.6.0), limma (>= 3.63.6) [C9] script uses vQuant/voomQuantiles output equal to deposited normalised matrix (section 2): max diff: 4.97e-14 [C9] STRTprep pipeline (conf.yaml, licence, pinned version): not part of the OSF deposit; not checked here [C10] files in the study deposit: 4 | README / licence / codebook / dictionary file names 0 | phenotype or SPSS files (.sav .sps .csv .xls*) 0 [C10] deposit contains: Table 1 data, cytokines, enrichment input/output, figure code, Table S1: none of these (only the limma step and its inputs) ==== C11. Sample codes (counts only, no code or per-child value printed) ==== [C11] distinct codes in the sample table: 27 | same codes in the headers of the counts and the normalised matrix TRUE, TRUE | code occurrences across the three files 81 [C11] distinct numeric stems: 25 [C11] children unique on: sex: 0 | group 0 | sex + group 0 | offset 23 | sex + offset 25 | sex + group + offset 25 [C11] offset rounded to half years: unique on sex + group + offset: 10 | whole years 4 | sex + whole-year offset 0 [C11] half-year offset refit, current default: shipped coefficient hits: 268 | shared with the 273 266 | additive hits 0 ==== Session ==== total run time (minutes): 11.9 R version 4.5.1 (2025-06-13 ucrt) Platform: x86_64-w64-mingw32/x64 Running under: Windows 11 x64 (build 26300) Matrix products: default LAPACK version 3.12.1 locale: [1] LC_COLLATE=English_United States.utf8 LC_CTYPE=English_United States.utf8 LC_MONETARY=English_United States.utf8 [4] LC_NUMERIC=C LC_TIME=English_United States.utf8 time zone: Europe/Prague tzcode source: internal attached base packages: [1] parallel stats graphics grDevices utils datasets methods base other attached packages: [1] edgeR_4.8.2 limma_3.66.0 loaded via a namespace (and not attached): [1] compiler_4.5.1 tools_4.5.1 splines_4.5.1 grid_4.5.1 locfit_1.5-9.12 lattice_0.22-7 statmod_1.5.2