Hi there

I have an issue with using gaussNorm on a flowSet constructed by using read.flowSet on .FCS files.

I can visualize the distributions of individual flowFrames using ggcyto, but whenever I run gaussNorm on the flowSet, I get the following error:

"Error in density.default(na.omit(lms.list)) : need at least 2 points to select a bandwidth automatically"

The following code reproduces the problem:

R version 4.1.2 (2021-11-01) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows 10 x64 (build 19045)

library(flowCore)
library(flowStats)
library(ggcyto)

data(ITN)

pITN <- ITN[1:2]

gaussNorm(pITN, "CD8", )$flowset

With the ITN data, I can fix the problem by choosing more elements for the pITN subset, or by transforming the data:

tpITN <- transform(pITN, "CD4"=asinh(CD4), "CD3"=asinh(CD3), "CD8"=asinh(CD8))

tpITN <- gaussNorm(tpITN, "CD8", )$flowset

However, none of those solutions work with my model data.

The following code visualizes the individual flowFrames of the flowSet, and also works on my model data:

#For the raw data
ggcyto(pITN, aes(x = "CD8", col = name)) + geom_density() + facet_null()

#For the transformed data
ggcyto(tpITN, aes(x = "CD8", col = name)) + geom_density() + facet_null()

Why does the error occur, despite the pITN dataset having 2 flowFrames?

How come the above solutions work with the ITN data?

And most importantly, what could I try to fix my issue, since the above solutions doesn't work for my model data?

Thank you in advance :-)

Mikkel