Error in rowSums(Ma) : 'x' must be an array of at least two dimensions

[nfancy]

Hi,

Thank you for this package. I have been looking for something like this. I tried to use this with my data but I got the following error.

Error in rowSums(Ma) : 'x' must be an array of at least two dimensions

My codes are:

egm = create_ENIGMA(bulk = as.matrix(countdata_all), 
                    ref = sce,
                    ref_type = "single_cell", 
                    meta_ref = colData(sce))

egm = batch_correct(egm, 
                    varname_cell_type = "celltype_global",
                    n_pseudo_bulk=100)

Here, I am using a SingleCellExperiment object. Can you please help with the error?

Nurun

[WWXkenmo]

Hi Nurun!
Thx for using our tool!
I have run the following code on my machine, it seems that the 'batch_correct' function could run on my data without error or warnings.

sce = SingleCellExperiment(assays = list(counts = exprs(ref_sc_sub)))
sce$celltype_global  = pData(ref_sc_sub)$main_celltype
egm = create_ENIGMA(bulk = Bulk, ref = sce, ref_type = "single_cell",meta_ref = colData(sce))
egm = batch_correct(egm, 
                     varname_cell_type = "celltype_global",
                     n_pseudo_bulk=10) # for saving time I just generate 10 pseudo-bulk samples
Sat Dec 25 11:32:53 2021 Reference is from Single Cell RNA-seq, doing batch correction in S mode. 
Sat Dec 25 11:32:53 2021 Generating pseudo bulk... 
Sat Dec 25 11:32:54 2021 Doing ComBat... 
Found 1133 genes with uniform expression within a single batch (all zeros); these will not be adjusted for batch.
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data

Sat Dec 25 11:32:55 2021 Restore reference... 

you could check our data right here to see any different with your data. Our data and run example is right here (data: https://github.com/WWXkenmo/ENIGMA/tree/master; code: https://htmlpreview.github.io/?https://github.com/WWXkenmo/ENIGMA/blob/master/vignettes/ENIGMA_toy2.html) I think the following point you need to pay attention

1. Do your sce object have sample id (cell barcode infor)

head(colData(sce))
DataFrame with 6 rows and 1 column
                       celltype_global
                    <character>
AAACCTGAGTAACCCT_13      B_cell
AAACCTGCAACGATGG_13     Myeloid
AAACCTGCATTAGCCA_13     Myeloid
AAACCTGGTCCATGAT_13      T cell
AAACCTGTCCTGTACC_13     Myeloid
AAACGGGCAAGAAAGG_13     Myeloid

2. Do your reference single cell data contains the major cell types but not includes any rare celltype (<1%)

table(colData(sce))/nrow(colData(sce))
    B_cell         EC      Fibro    Myeloid     T cell      Tumor 
0.11551043 0.03350254 0.05561196 0.22470389 0.41534123 0.15532995 

I hope you could return those information to me or you could give me a toy data (subset of your dataset) as I can't see the details of your dataset to help me fix the potential bugs.

Thx again for using our tool, any suggestions are valuable and welcome!

Best,
Ken

[nfancy]

Hi, Thanks for your response. I've been looking at the code and I think the source of the error is in the R function S_mode_correction.R between line 46-50. Perhaps after normalisation in a row it becomes all zero again. I know that does not sound likely, but after adding this line frac <- frac[rowSums(frac)>0,] after normalization, it seems to have solved the error. If I encounter it again, I'll send you a toy data.

Nurun