Comparing differential expression across samples - is batch effect correction needed?

Question

I have a bulk RNA-seq dataset made up of control and treatment conditions for a range of cell lines. This dataset was generated in two batches, such that the cell lines are split between batches but both the treatment and control for each cell line are within the same batch. As the cell lines are very different, I'm looking to do DE analysis for each cell line and then compare these DE responses between cell lines.

I'm wondering whether batch effect correction is useful here since I'm only ever comparing DE values and not the raw counts themselves. Also, the baseline expression values for the cells lines are so different that they already constitute a strong biological batch effect that is perfectly correlated with the technical batch effect, which batch correction wouldn't be able to account for.

Answer 1

There is no need for batch correction because all your comparisons are within batch. As long as you are making comparisons for each line, and treatment and control for the same line are in the same batch, then any batch effects are already accounted. for.

Answer 2

If your treatment and control are prepared in different batches, it's impossible to distinguish between treatment-specific effects, and batch-specific effects. DESeq2 will complain when you attempt to carry out such correction in the statistical model for this reason.

Baseline differences in expression are a large reason for carrying out batch correction in the first place; it makes differential expression analyses more powerful, not less. I'd recommend consulting with a statistician (prior to carrying out an experiment / sequencing) regarding the best way to split samples into different batches so that batch effects can be discovered and accounted for. As you mention, there are multiple effects that should be considered (e.g. cell lines, sequencing run).