error message from DESeq2/EdgeR

usegalaxy.org support
,
1

Hi,

I have encountered a problem with Galaxy analyses of RNAseq data when I want to analyze the differential expression using DESeq2 or EdgeR. Below are the typical error messages. I have used HISAT2 for as the alignment program, Stringtie for assembly, and both htseq count and featurecounts to count the reads. I have tried to use the output from either htseq count, featurecounts and Stringtie into DESeq2 or EdgeR with error messages popping up after a few minutes. The original data are from Illumina next seq, paired end. I am working with Galaxy Main https://usegalaxy.org.

Below is from DeSeq2

Fatal error: An undefined error occurred, please check your input carefully and contact your administrator. Error in data.frame(…, check.names = FALSE) : arguments imply differing number of rows: 51

Below from EdgeR

Fatal error: Exit code 1 () Error in data.frame(…, check.names = FALSE) : arguments imply differing number of rows: 51329, 58444, 53946 Calls: do.call -> cbind -> cbind -> data.frame

2

Check the header , your input file must had it or you got to turning that off

3

Hi,

Thank for the response. I have tried turning off header option as well but still get error messages. Please see info below.
The data are from Illumina NextSeq 40 bp paired, TruSeq stranded RNA prep kit used

With header option selected:

Fatal error: An undefined error occurred, please check your input carefully and contact your administrator.

Error in data.frame(…, check.names = FALSE) :

arguments imply differing number of rows: 4, 58443

Calls: get_deseq_dataset … eval -> eval -> eval -> .Method -> cbind -> data.frame

With header option turned off:

Fatal error: An undefined error occurred, please check your input carefully and contact your administrator.

Error in Ops.factor(a$V1, l[[1]]$V1) :

level sets of factors are different

Calls: get_deseq_dataset … DESeqDataSetFromHTSeqCount -> sapply -> sapply -> lapply -> FUN -> Ops.factor

Warning message:

In is.na(e1) | is.na(e2) :

longer object length is not a multiple of shorter object length

Output HISAT2 sample

4

It looks like your data does not have a header (where the first error comes from). This can be inspected to confirm (no need to guess) by clicking on the “eye” icon to view the first lines of one (or all) of the count datasets. Galaxy will create headers if none are included already in the inputs.

This sometimes comes up (second error) when the wrong input is selected by mistake. Example: summary/report data was input instead of count data.

If you would like to compare to sample data/workflows, please see the RNA-seq tutorials. “Reference-based RNA-Seq data analysis” includes DESeq2.

5

if your header was correct… maybe you make a mistake in loading your data in Deseq2 or EdgeR. you cant use a data twice in one run of Deseq2 or EdgeR. i think you do something wrong in loading Steps.take a Sc of your analyse data

1 Like
6

Hello,

I’ve also encountered the same problem with deseq2 using htseq counts. I’ve made sure the header options are turned off and each dataset is used once. Is there anything else I should consider or look out for when loading my data?

Error in Ops.factor(a$V1, l[[1]]$V1) :
level sets of factors are different
Calls: get_deseq_dataset … DESeqDataSetFromHTSeqCount -> sapply -> sapply -> lapply -> FUN -> Ops.factor