Missing channels in Phenograph cluster

[laufre]

Hi,
I'm experiencing the issue, that if all cells in one sample have the value = 0 for a channel, this channel cannot be selected for e.g. Phenograph.
Is there a workaround, allowing me to select a channel for Phenograph, even though one sample out of many is completely = 0 for this marker?
(essentially the same issue mentioned here: http://imc-forum.org/viewtopic.php?f=4&t=98)
However, in my case all cells being zero for a given marker is not due to bad data quality but intended.

Your help would be very much appreciated!

[DenisSch]

Unfortunately, there is currently no option to allow this, since this sample will create an issue when running PhenoGraph. As a workaround you could add a very small number to all samples to prevent channels from being zero?

[laufre]

I did this as a workaround and it helps, but now I face a different problem:
If I try to merge samples or draw a gate and have a look at the newly added Population, all markers have really weird values, both negative and positive and really high (see attached comparison between all samples before merging and afterwards for the same marker).
I'm unsure what causes this, there is no error message, only this:

File loaded: 0.326998s
Allocated space for data: 0.738058s
Read data into session: 0.869827s