Closed denvercal1234GitHub closed 1 year ago
Hello,
If you used a non-default path_to_intermediary_results
argument, that folder will contain an rds
subfolder which contains a file called xp_transformed.rds
. This contains the logicle-transformed data for all events in matrix format. Maybe you could use it to plot histograms and make sure the transformed data distribution looks satisfactory ?
Wish you the best for your revision
Thank you again very much, @ebecht.
For visualization, likely just the transformed, but is there a case when we should use the xp_transformed_scaled.Rds
instead of the xp_transformed.Rds
?
The scaling is used before models training to harmonize data across wells, I don't think it should be used for other purposes unless you think it is a good idea to do so
Hi there,
Thank you again for the package.
A reviewer asked to confirm whether we checked whether the logicle transformation was done accurately by visualizing the marker expressions on density map for example.
But, the way the
infinity_flow()
function is written, there is not an intermediate step to check this accuracy.Would you mind advising me how we coud still check this potential issue?
Thank you for your help.
Related to issue #6.