Rounding Proseg Expected Counts

[annachernysheva179]

Hi,

Thank you for developing this tool! I'm using it for the segmentation of Xenium data, followed by classical single-cell analysis. However, I've noticed that the expected counts from Proseg are not integers, whereas single-cell raw counts typically are. Do you have any recommendations on the impact of rounding Proseg expected counts? Would rounding them result in a loss of important information?

Best, Anna

[dcjones]

Hi Anna,

Rounding the nearest integer should be completely fine. I do that pretty regularly when a downstream analysis assumes integer counts.