I have a question regarding Batch and Covariates in ComBat. I have an Illumina expression dataset of control versus treatment, at 3 different time points (day3,5,7) and 4 biological replicates each. However RNA from different experimental groups (replicates) was extracted at different time points and by default samples cluster together based on the RNA extraction date therefore I have a strong batch effect. Should I use RNA extraction date as Batch and both the biological replicate groups and Time_x_Treatment as covariates (like the sample file is shown below) or apply it in any other way (using extraction date as first batch and then re-run with replicate groups as second batch)? Second question, is it correct to subgroup the covariate as Time_x_Treatment or should I just have control vs treatment (and not specify the day factor)? Array Sample Batch Covariate Covariate2 D3C4 day3control_repl4 1 day3control 1 D3C3 day3control_repl3 2 day3control 3 D3C2 day3control_repl2 3 day3control 4 D3C1 day3control_repl1 3 day3control 2 D5C4 day5control_repl4 2 day5control 3 D5C3 day5control_repl3 1 day5control 1 D5C2 day5control_repl2 1 day5control 4 D5C1 day5control_repl1 4 day5control 2 D7C4 day7control_repl4 5 day7control 4 D7C3 day7control_repl3 4 day7control 3 D7C2 day7control_repl2 2 day7control 2 D7C1 day7control_repl1 1 day7control 1 D3T4 day3treatment_repl4 1 day3treatment 1 D3T3 day3treatment_repl3 4 day3treatment 3 D3T2 day3treatment_repl2 1 day3treatment 4 D3T1 day3treatment_repl1 6 day3treatment 2 D5T4 day5treatment_repl4 5 day5treatment 3 D5T3 day5treatment_repl3 4 day5treatment 1 D5T2 day5treatment_repl2 6 day5treatment 4 D5T1 day5treatment_repl1 6 day5treatment 2 D7T4 day7treatment_repl4 2 day7treatment 4 D7T3 day7treatment_repl3 4 day7treatment 3 D7T2 day7treatment_repl2 5 day7treatment 2 D7T1 day7treatment_repl1 2 day7treatment 1 -- output of sessionInfo(): R version 2.15.2 (2012-10-26) Platform: i386-apple-darwin9.8.0/i386 (32-bit) locale: [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8 attached base packages: [1] splines stats4 graphics grDevices utils datasets grid stats methods base other attached packages: [1] mgcv_1.7-22 corpcor_1.6.4 RColorBrewer_1.0-5 panp_1.28.1 hwriter_1.3 R2HTML_2.2 reshape_0.8.4 plyr_1.7.1 gProfileR_0.2 fpc_2.1-5 [11] flexmix_2.3-8 multcomp_1.2-14 survival_2.36-14 mvtnorm_0.9-9993 modeltools_0.2-19 lattice_0.20-10 mclust_4.0 MASS_7.3-22 cluster_1.14.3 preprocessCore_1.20.0 [21] affy_1.36.0 Biobase_2.18.0 BiocGenerics_0.4.0 limma_3.14.1 pheatmap_0.7.4 gridExtra_0.9.1 ggplot2_0.9.2.1 -- Sent via the guest posting facility at bioconductor.org.