Mapping of ENSG_IDs between different release of the Ensembl database

[Olga013]

Dear members of the Ensembl team,

I wasn’t sure who to contact, so I’m starting here.

I am writing to ask you questions about the IDMapper tool presented on your website: https://www.ensembl.org/Homo_sapiens/Tools/IDMapper. I am working with the single-cell gene expression data and would like to map the outdated ensembl_ids corresponding to the 90/91 releases of the Ensembl database to the ids from the recent release of the database (115).

For example, I ran the query with the following IDs:
ENSG00000251678
ENSG00000281557
ENSG00000251931

It gave me the results represented in the file, which I attached to this request.
Additionally, I ran BioMart (https://www.ensembl.org/info/data/biomart/index.html) and also attached the results to the file.

I have several questions:

• Why does IDMapper give the mapping ENSG00000251678 → ENSG00000251678? According to the search engine in the Ensembl service, it is written that the gene with the id ENSG00000251678 is remapped to another id: https://www.ensembl.org/Homo_sapiens/Gene/Idhistory?g=ENSG00000251678

• Why does ENSG00000281557 have the mapping as ENSG00000281557 → ENSG00000281557 ? According to the search engine (as well as gene.txt.gz, https://ftp.ensembl.org/pub/release-115/mysql/homo_sapiens_core_115_38/gene.txt.gz, from the 115 release), there is no gene with such an ens_id.

• Why is there no such mapping as ENSG00000251931 → ENSG00000251931? On the other hand, we can easily find the related gene by searching through this ens_id: https://www.ensembl.org/Homo_sapiens/Gene/Summary?db=core;g=ENSG00000251931;r=12:59450673-59450772;t=ENST00000516122 ? Furthermore, when I ran the query only with ENSG00000251931 it also gave me no results. In contrast, when I ran the BioMart query, ENSG00000251931 is the only gene_id that was found.

• Additionally, I went through mapping_session file: https://ftp.ensembl.org/pub/release-115/mysql/homo_sapiens_core_115_38/mapping_session.txt.gz and saw in the column new_release the value 114 in the last row, but new_db_name contain homo_sapiens_core_115_38 with the value 115. I am a bit confused. Should it be 114 or 115 in the new_release column in the last row?

I am looking forward to your response.

Thank you in advance!

idmapper.pdf

@szalata

[sgiorgetti]

Hello @Olga013 - thanks for bringing this up.

It looks like we've got a few issues here.

• ENSG00000251678: it seems to have been merged into ENSG00000223597 in release 113. However, the result of the ID history converter tool shows it being current (115).
• ENSG00000281557: its history seems to stop at release 100, while in reality the gene goes happily on until release 109. In release 110 it disappears. Its vanishing may be due - still to confirm - of it being a gene sitting across a patch boundary; hence ending up being removed. Annoyingly, there is no trace in the history.
• ENSG00000251931: this gene has not changed since release 58, and it's current. A mapping event should be logged, when a change (maybe "only" in the version) is detected.
• Finally, the "new_release" in row 425 of mapping_session table is incorrect: it should be 115, instead of 114.

We'll be amending data in release 116, which is already in advanced state of processing and due 1Q26.

Thanks again for flagging these issues up, and for using our services.