A rather large amount of entries contain a whole lot of genes with identical expression levels, especially some that should be vastly different, i.e. HLA-A and HLA-B.

Here's an example after fetching data using depmap_copyNumber(): depmap_id: ACH-000038 log_copy_number: 1.004462

which is found in 2550 different genes.

Here's another one:

I think I might be missing something here, could anyone help?

This was provided as an answer by email to OP. I copy it here so that other benefit from it:

We did not change any of these copy number values. Integer values in Depmap R package datasets are as-is. The genes with similar CNV counts are likely an artifact of normalization from their pipeline, which you can read more about here: https://depmap.sanger.ac.uk/documentation/datasets/copy-number/

If you have more specific questions about how this data was normalized, I suggest that you address your questions to depmap@sanger.ac.uk