Help with manual inspection of AUCell histogram for regulon binerization

[Jeron1996]

Hi,

Thanks for developing this great package!

I want to binarize my regulons. It's advised to manually inspect the threshold that was automatically generated. I wanted some input on this manual inspection.
Currently, I am comparing the histogram to the chosen distribution and evaluate whether the histogram follows this distribution. For most cases that seems to work. However, occasinally the histogram falls between two distributions (see example below). Most often between the global distribution (gray) or the L_k2 (red) distribution. What do you advise to do in these cases? Keep the automatically chosen distribution or change it to the other one?

Thanks in advance!