Batch-Corrected Counts from RPCA with log-normalized counts

[andrewdchen]

Similar question to the one raised in 8146 but for log normalized data.

But running integration with IntegrateLayers as demonstrated in the vignette, doesn't return a batch-corrected counts layer if you already log-normalized your data. I'm running code like the below:

obj <- IntegrateLayers(
  object = obj, method = RPCAIntegration,
  orig.reduction = "pca", new.reduction = "integrated.rpca",
  verbose = FALSE
)

and there's a layer called 'integrated.rpca' with a (Nx50) (50 for the #PCA dims in my 'pca' reduction). But there's no corrected counts that is (Nxp) (p being the number of genes), that would have been available if I used the v3 function IntegrateData.

Is there a way to use the v5 function whilst still outputting batch corrected counts?