Hi, I have some outdated Ensembl Gene IDs. I tried using Ensembl's ID converter tool to get the new versions of the IDs but it just spits back the same ID to me that wasn't working in the first place. For example, with "ENSGALG00000015646", the tool says that the ID is still in the current release, Release 90 (see below). But if I type the ID into the search bar on the website, it gives 0 results. Any idea what's going on?
This is strange behaviour- and I think you may have found a bug in the ID history convertor. I think the problem arises from where we updated form the Galgal4 assembly to the Gallus_gallus-5.0 assembly.
Obviously, this is a long-way round to get the result you are looking for with the ID history convertor, so we are looking into fixing this issue as quickly as possible. I will keep you updated with the progress of our work.
Having looked into this problem- it's caused by the way that retired Ensembl stale IDs are labelled in our database. This means that the ID history convertor does not recognise retired stable IDs and simply lists the retired stable ID for the current Ensembl release.
This is only a problem where an old stable ID is not linked to a new stable ID. Where an old stable ID is linked to a stable ID in the current version of Ensembl, the ID history convertor will be able to provide an accurate history.
With ENSGALG00000015646, you have found an example where the ID is retired and not mapped to a gene in the current version of Ensembl. This happens in the instances where the mapping pipeline does not detect a sufficiently similar gene.
To fix this, we will change the way stable IDs are retired in our database, so that the results the ID history convertor provides for retired stable IDs is not ambiguous. However, we are unable to change the pipeline that maps the stable IDs, and so from Ensembl 91 onwards you will see ENSGALG00000015646 simply listed as 'retired'.
I hope this helps clarify the data you retrieved form the ID history convertor, but please do get back in touch if anything is still unclear.
Hi,
This is strange behaviour- and I think you may have found a bug in the ID history convertor. I think the problem arises from where we updated form the Galgal4 assembly to the Gallus_gallus-5.0 assembly.
For example, searching in the Ensembl archives (in a version that used the Galgal4 assembly) provides an entry for your stable ID: http://dec2015.archive.ensembl.org/Gallus_gallus/Gene/Summary?db=core;g=ENSGALG00000015646;r=2:117721181-117734940;t=ENSGALT00000025230
You can then use the gene name (TCEB1) to find the gene in the current version of Ensembl: http://www.ensembl.org/Gallus_gallus/Gene/Summary?db=core;g=ENSGALG00000035417;r=2:118445078-118458839;t=ENSGALT00000065415
Obviously, this is a long-way round to get the result you are looking for with the ID history convertor, so we are looking into fixing this issue as quickly as possible. I will keep you updated with the progress of our work.
Best wishes
Ben Ensembl Helpdesk
This gene is indeed not in e90 but present in older versions. What ID conversion tool are you using ?
The ID History Converter tool on Ensembl's webite: http://www.ensembl.org/Homo_sapiens/Tools/IDMapper?db=core Do you have a different one you'd recommend?
I did not check myself, but Jean-Karim is confirming .... so maybe tell them that their converter is not working, there might be a bug!
This looks like a bug to me. You should report to the helpdesk.