Closed Lan-h closed 1 year ago
NB: I tried to "normalize" the thresholds based on sequencing depth so as to get a constant top number of genes, but it is not working on all datasets so I wondered if there was another parameter to be taken into account to be able to compare datasets.
Sorry, for questions on how to use the package, see: https://github.com/aertslab/AUCell/issues/26
Hi ! I would like to use AUCell on different datasets with different sequencing depth. Some of them are already integrated so I guess another issue is that 0 have also been added in the count matrices.
I wonder if there is a way to take this into account and set different thresholds to offset the different sequencing depth and be able to compare AUC scores across datasets.
Thanks in advance for your help !