Entering edit mode
dge$samples$treatment <- factor(dge$samples$treatment, levels = c("WT", "KO"))
dge$samples$timepoint <- factor(dge$samples$timepoint, levels = c("bl", "day4", "day14"))
design <- model.matrix(~ batch + treatment * timepoint, data = dge$samples)
contr.mat <- makeContrasts(
treatment_KOvsWT = treatmentKO,
time_day4vsbl = timepointday4,
time_day14vsbl = timepointday14,
interaction_day4 = treatmentKO.timepointday4,
interaction_day14 = treatmentKO.timepointday14,
KO4_vs_WT4 = treatmentKO + treatmentKO.timepointday4,
KO14_vs_WT14 = treatmentKO + treatmentKO.timepointday14,
levels = design
)
> sessionInfo()
R version 4.5.0 (2025-04-11 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 11 x64 (build 22621)
Matrix products: default
LAPACK version 3.12.1
locale:
[1] LC_COLLATE=German_Germany.utf8 LC_CTYPE=German_Germany.utf8
[3] LC_MONETARY=German_Germany.utf8 LC_NUMERIC=C
[5] LC_TIME=German_Germany.utf8
time zone: Europe/Berlin
tzcode source: internal
attached base packages:
[1] stats graphics grDevices utils datasets methods base
loaded via a namespace (and not attached):
[1] mnormt_2.1.1 deldir_2.0-4 pbapply_1.7-4 gridExtra_2.3
[5] rlang_1.1.6 magrittr_2.0.3 RcppAnnoy_0.0.22 matrixStats_1.5.0
[9] ggridges_0.5.6 compiler_4.5.0 spatstat.geom_3.5-0 png_0.1-8
[13] vctrs_0.6.5 reshape2_1.4.4 stringr_1.5.1 pkgconfig_2.0.3
[17] fastmap_1.2.0 promises_1.3.3 purrr_1.1.0 jsonlite_2.0.0
[21] goftest_1.2-3 later_1.4.2 spatstat.utils_3.1-5 psych_2.5.6
[25] irlba_2.3.5.1 parallel_4.5.0 cluster_2.1.8.1 R6_2.6.1
[29] ica_1.0-3 stringi_1.8.7 RColorBrewer_1.1-3 spatstat.data_3.1-6
[33] limma_3.64.3 reticulate_1.43.0 parallelly_1.45.1 spatstat.univar_3.1-4
[37] lmtest_0.9-40 scattermore_1.2 Rcpp_1.0.14 tensor_1.5.1
[41] future.apply_1.20.0 zoo_1.8-14 sctransform_0.4.2 httpuv_1.6.16
[45] Matrix_1.7-3 splines_4.5.0 igraph_2.1.4 tidyselect_1.2.1
[49] abind_1.4-8 codetools_0.2-20 spatstat.random_3.4-1 miniUI_0.1.2
[53] spatstat.explore_3.5-2 listenv_0.9.1 lattice_0.22-6 tibble_3.2.1
[57] plyr_1.8.9 shiny_1.11.1 ROCR_1.0-11 Rtsne_0.17
[61] future_1.67.0 fastDummies_1.7.5 survival_3.8-3 polyclip_1.10-7
[65] fitdistrplus_1.2-4 BiocManager_1.30.26 pillar_1.11.0 Seurat_5.3.0
[69] KernSmooth_2.23-26 plotly_4.11.0 generics_0.1.4 RcppHNSW_0.6.0
[73] sp_2.2-0 ggplot2_3.5.2 scales_1.4.0 globals_0.18.0
[77] xtable_1.8-4 glue_1.8.0 lazyeval_0.2.2 tools_4.5.0
[81] data.table_1.17.8 RSpectra_0.16-2 locfit_1.5-9.12 RANN_2.6.2
[85] dotCall64_1.2 cowplot_1.2.0 grid_4.5.0 tidyr_1.3.1
[89] edgeR_4.6.3 nlme_3.1-168 patchwork_1.3.1 cli_3.6.5
[93] spatstat.sparse_3.1-0 spam_2.11-1 viridisLite_0.4.2 dplyr_1.1.4
[97] uwot_0.2.3 gtable_0.3.6 digest_0.6.37 progressr_0.15.1
[101] ggrepel_0.9.6 htmlwidgets_1.6.4 SeuratObject_5.1.0 farver_2.1.2
[105] htmltools_0.5.8.1 lifecycle_1.0.4 httr_1.4.7 statmod_1.5.0
[109] mime_0.13 MASS_7.3-65
I originally tried to include my full question and code in one post, but there was an upload problem. I am therefore posting the code separately above, and here in this comment I am adding the questions and experimental context.
I have an scRNA-seq experiment with mice and two factors:
treatment: WT / KO
timepoint: baseline (bl), day4, day14
Mice were sacrificed at each timepoint (no repeated measures). We also have a batch variable (sequencing batch).
Raw counts (after QC) were pseudobulked per subcell type, for each combination of condition, timepoint, and batch (e.g., capillary endothelial cells, KO, day4, batch 7).
Goal:
1) Test KO vs WT within each timepoint (bl, day4, day14) 2) Test for an interaction (KO-WT difference changes from baseline to day4/day14) 3) Keep batch as a covariate
Questions:
1) Are my contrasts correct for KO vs WT within day4 and day14? 2) Can I use these "direct comparisons" without the interaction term if I only care about KO vs WT at a single timepoint? 3) Should I keep batch in the model even after Harmony batch correction during clustering?