Dosing not identical between CP and L1000

[Lawrence-Phillips]

From the Haghighi et. al. paper from 2022 describing the Rosetta data: Each plate is treated with chemical or genetic perturbations identically.
But upon closer examination of the actual data files, this does not appear to be the case, at least for CDRP.

For compound BRD-K18316707-001-01-9:

L1000 data (replicate_level_l1k.csv): 10.0 micromolar dose
Cell Painting data: 5.00 micromolar dose
For compound BRD-K18250272-003-03-7:

In the L1000 data(replicate_level_l1k.csv): 6.05 micromolar dose
In the Cell Painting data: 3.02 micromolar dose

It seems that while the same compounds are used, the doses are not identical between the L1000 and Cell Painting assays - the L1000 doses appear to be roughly double the Cell Painting doses in these examples. Is this a typo, if so, which one is the right dose?

[Lawrence-Phillips]

To follow up this is me using the bulk annotations found here (not CP features):

https://idr.openmicroscopy.org/webclient/annotation/35793307