ArchR multiome analysis: varFeature in scRNA-seq addIterativeLSI()

[Jiahuan90]

Hi rcorces,

I met a problem that I ran multiome ArchR pipeline scRNA-seq part with varFeature = 2500 successful.
proj <- addIterativeLSI(
ArchRProj = proj,
clusterParams = list(
resolution = 0.2,
sampleCells = 10000,
n.start = 10
),
saveIterations = FALSE,
useMatrix = "GeneExpressionMatrix",
depthCol = "Gex_nUMI",
varFeatures = 2500,
firstSelection = "variable",
binarize = FALSE,
name = "LSI_RNA"
)

However, when I adjust the varFeatures = 25000, it gave me error messages: "nFeature for variable selection must be at leat 1/2 the total features!".

I don't know what is the meaning of varFeatures and how to set the rationale number for this parameter in scRNA-seq part? 2500? 5000? or something else?

My second question is: should the varFeature parameters be same in scATAC-seq part? Since I use varFeature = 25000 in scATAC-seq part successful.

Thank you for your kindly help!

[rcorces]

Sorry. That is a typo which I've corrected via 21099d6

It should say "nFeature for variable selection must be less than 1/2 the total features"

You cannot have varFeatures be greater than half of the total features. Because ATAC has way more peaks than RNA has genes, you should not use the same values for varFeatures in both cases.

[Jiahuan90]

Thank you for your prompt and helpful response!