Do you have nonetheless an idea what it could mean? I am surprised that there is no info about this in the internet. Like I said I just downloaded the GTF and genome FASTA from https://www.gencodegenes.org/human/release_33lift37.html , which is widely used to my knowledge. Do you think the GTF file is in some way damaged? Or is it save for me to work with this release (GRCh37, Release 33)?
Hello everybody,
I performed a DGE analysis and now I am trying to convert the gene ID to gene names.
The problem is that virtually all of my gene IDs look somewhat like this:
ENSG00000000003.15_4.
In the list of gene IDs which correspond to the gene name, the gene IDs have this format:
ENSG00000000003.15
I used kallisto to quantify my reads and the genome as well as the GTF file from GENCODE (https://www.gencodegenes.org/human/release_33lift37.html)
Can I just delete everything after the _ and proceed? And what is the meaning of the underscore in the gene ID in general?
Any help is appreciated! Thanks!
1 answer
Yes, Kevin and ATpoint are correct. The underscore is not part of the Ensembl stable ID specifications.
Best wishes
Ben Ensembl Helpdesk
Ah, sorry - I didn't notice the files from your original post. In the GENCODE GTFs mapping versions are appended to the identifiers (eg. ENSG00000228327.3_2) in the annotation mapped back to GRCh37.
More information can be found in their documentation: https://www.gencodegenes.org/pages/data_format.html
Log in to answer this question.
Hmm, never came across underscores using GENCODE. Maybe this has to do with the lift from hg38 to hg19. I personally even remove the version numbers, so this probably does not do any harm.
Indeed, it is not part of the official format specification: https://www.ensembl.org/Help/Faq?id=488
You could contact the Ensembl Help Desk.