Getting Blank result

[ShaktiprasadM]

I am using TCGA-LGG data. I used the following code but getting blank output.
I have saved genes and probesets under genes and probesets.
MCPcounter.estimate(rna, featuresType = "ENSEMBL_ID", probesets = probesets, genes = genes)

[ebecht]

Hi,

Sorry I missed your post. In case it's not too late, for diagnosis could you print rownames(rna) and rna[1:5, 1:5] ?

[Donbbit]

I have the same issue, do you know how to solve it?

[ebecht]

Hi,

Sorry I missed your post. In case it's not too late, for diagnosis could you print rownames(rna) and rna[1:5, 1:5] ?

Start with this ?

[rebelipar]

Hello!

I am having the same issue.

This (where TCGA_counts is a df with ENSEMBL IDs as rownames and columns of sample IDs) only returns an empty matrix.

TCGA_MCP = MCPcounter.estimate(TCGA_counts, featuresType="ENSEMBL_ID")

To answer the questions you asked for others:
rownames(TCGA_counts) returns this (well it goes on much longer, but this a piece of it):
[1] "ENSG00000000005.6" "ENSG00000000419.13" "ENSG00000000457.14" "ENSG00000000460.17" [5] "ENSG00000000938.13" "ENSG00000000971.16" "ENSG00000001036.14" "ENSG00000001084.13" [9] "ENSG00000001167.14" "ENSG00000001460.18" "ENSG00000001461.17" "ENSG00000001497.18" [13] "ENSG00000001561.7" "ENSG00000001617.12" "ENSG00000001626.16" "ENSG00000001629.10" [17] "ENSG00000001630.17" "ENSG00000001631.16" "ENSG00000002016.18" "ENSG00000002079.14" [21] "ENSG00000002330.14" "ENSG00000002549.12" "ENSG00000002586.20" "ENSG00000002586.20_PAR_Y" [25] "ENSG00000002587.10" "ENSG00000002726.21" "ENSG00000002745.13" "ENSG00000002746.15" [29] "ENSG00000002822.15" "ENSG00000002834.18" "ENSG00000002919.15" "ENSG00000002933.9" [33] "ENSG00000003056.8" "ENSG00000003096.14" "ENSG00000003137.8" "ENSG00000003147.19" [37] "ENSG00000003249.15" "ENSG00000003393.15" "ENSG00000003400.15" "ENSG00000003402.20"

and TCGA_counts[1:5, 1:5] returns this:
TCGA.35.5375.01A TCGA.55.A4DF.01A TCGA.MP.A4T4.01A TCGA.62.A471.01A TCGA.L9.A5IP.01A ENSG00000000005.6 0 0 0 0 0 ENSG00000000419.13 2468 3162 1587 1071 1180 ENSG00000000457.14 249 726 747 630 982 ENSG00000000460.17 395 603 370 542 642 ENSG00000000938.13 150 1578 1634 372 361

I also get a blank matrix if I import the counts data without using the gene IDs as rownames. In that case, rownames just returned numbers.

I am currently using R 4.4.1. And I am on Windows if that matters.

[ebecht]

Hi @rebelipar
I think this is because your rownames include version numbers of the gene annotation, while gene identifiers in our signatures do not.

Can you try the following and see if that fixes your problem ?
rownames(TCGA_counts) = sapply(strsplit(rownames(TCGA_count), ".", fixed = TRUE), "[", 1)

Best
Etienne

[rebelipar]

Ha, that totally fixed it, thank you! I had to deal with some duplicate first, but it is working!