get gene ids running sleuth kallisto for differential gene expression

Question

[frida.danielsson]

Hi,

I'm trying to run sleuth to look for differential gene expression between two conditions, but I find the manual a bit hard to interpret.

After running kallisto I have extensions in the transcript IDs (like "ENST00000632859.1") and I assume that's why I cannot get the output data mapped to gene IDs while running biomaRt as described in the "getting started" webpage, anyone knows how to deal with this? Can I just remove the extensions in the .tsv file or do I need to change something in the bootstrap file as well?

[WouterDeCoster]

Did you use their transcriptome index? Because that one gives me nicely formatted name, causing no problem to map them to the gene names.

[frida.danielsson]

I used a fasta file for the human genome looking like this:

>ENST00000448914.1 havana_ig_gene:known chromosome:GRCh38:14:22449113:22449125:1 gene:ENSG00000228985.1 gene_biotype:TR_D_gene transcript_biotype:TR_D_gene
ACTGGGGGATACG...

Maybe I should try rebuild my index and rerun kallisto...

[WouterDeCoster]

These are the transcriptomes they suggest to use: http://bio.math.berkeley.edu/kallisto/transcriptomes/

Answer 1

[Devon Ryan]

Removing the .1 shouldn't be a problem. Whether it's there or not is due to whether your files came from Ensembl or Gencode (or maybe UCSC) originally.

[frida.danielsson]

Ok, so you mean I can keep it?

[Devon Ryan]

That means you can remove it if it's causing problems.