Hi

I'm trying to compare RNA Seq data from human tumors with mice primary cells. Is it possible to import it directly through tximeta ?

From the following code, I guess that this is because transcript ID are different, which is indeed the case...

se <- tximeta(coldata, type = "kallisto")
1 2 3 4 5 6 7 8 9 10 11 12 13 Error in tximport(files, type = type, txOut = txOut, ...) : 
  all(txId == raw[[txIdCol]]) is not TRUE
In addition: Warning message:
In txId == raw[[txIdCol]] :
  longer object length is not a multiple of shorter object length

If tximeta is not able to point to multiple references, is there a proper way to perform this analysis ? The quantification was performed using Kallisto.

Thanks a lot !

Cheers

Guillaume

I don't have any advice or automated workflows for cross-species analysis. One issue is that there isn't a cross-species container: SummarizedExperiment is really designed such that the seqnames comes from a single organism (you cannot have two chr1).

tximeta is designed for either reference transcriptomes or transcrpitomes you have defined with FASTA and GTF files as metadata.

Maybe it would be easier for you to use tximport to import the matrices, but then the actual analysis steps will be up to you.