RunCellHGT error

[lyuguol]

Dear CelliD developers,

I am using CelliD to analyze my own scRNA-seq data from 10X genomics. When running the 'RunCellHGT', I encountered the following errors:

Error in rep(seq(length(PathwayMat)), times = PathwayLen) : invalid 'times' argument

or

Error in pathways[sapply(pathways, function(x) length(x) >= minSize)] :
invalid subscript type 'list'

Do you know how these happen?

FYI, it is successful to use the data from the Vignette.

Thank you and best,
Guoliang

[Cortalak]

Hello, thanks for your feedback

for this error
Error in rep(seq(length(PathwayMat)), times = PathwayLen) : invalid 'times' argument

It is most certainly because the pathway argument was not a named list or the geneset size is too short and hence everything is filtered by the minSize argument.

• This can be due to a minSize value being too big. The default value is 10 and hence every geneset with less than 10 genes present in your single cell data will be removed and not considered.

• It can be a problem of nomenclature. Geneset being an ensemblID but the single cell data in GeneSymbol or genset having human genes but single cell data is from mouse.

Akira

[devashishmohanta]

Dear @Cortalak
I have been using mouse brain sc-RNA seq data to follow the Cell ID pipeline.
There was an issue with the nomenclature since the Panglao database has all the gene names/symbols given in CAPITALS so I tried capitalizing the gene names in my single Cell data to follow the CellID pipeline but it throughs the same error.
I even tried reducing the minSize to see if the error resolves but it throughs the same error again.
Please give any other insights to resolve the error.

Thanks and regards

[Cortalak]

Hello @devashishmohanta ,

That's very strange indeed, could you try something like
sapply(geneset, function(x) sum(x %in% rownames(SeuratObject)))

that should throw the number of genes in the geneset that is actually present in the data

[Cortalak]

For instance with a mock example

**library(Seurat)
library(CellID)
ngenes = sapply(Hallmark, function(x) sum(x %in% rownames(pbmc_small)))**       

You should have something like this

[dnyansagar]

Hi @Cortalak,
I am having the same error message.
I am attaching the screenshot of the geneset overlap with my data here.
any help would be appreciated

[TangentDeng]

@Cortalak Hi Akira, same issue. My samples are mice. So I set

panglao_pancreas <- panglao_pancreas %>%  filter(str_detect(species,"Mm"))

After run

ngenes = sapply(Hallmark, function(x) sum(x %in% rownames(Baron)))

for

sapply(geneset, function(x) sum(x %in% rownames(SeuratObject)))
Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'X' in selecting a method for function 'sapply': object 'geneset' not found

any help would be appreciated!