Hi,

While running DESeq2, I corrected for covariates in the design formula and extracted the vst matrix with blind = FALSE. If I understand it correctly from here (https://www.bioconductor.org/packages/devel/bioc/vignettes/DESeq2/inst/doc/DESeq2.html#blind-dispersion-estimation), blind = FALSE takes into account the design formula so is my resulting vst matrix already corrected for covariates? I am asking this because I want to do downstream analysis such as running single sample GSEA (ssGSEA) and was wondering if using this vst matrix as input will already take into account the covariates and I might not have to correct for them later. Please let me know!

Thanks, Manik

No the transformations do not correct for the design. See the FAQ in the vignette for more detail (also brought up in recent posts here).