EnsemblEnsembl Home

Ensembl Archives

About Archive Ensembl

Archive_ensemblThe main Ensembl site (www.ensembl.org) and the mirror sites are updated with the latest data approximately every three months. We maintain the Ensembl Archive sites so that there are stable links to data from a particular release. As of December 2016 these will be available for five years, together with the following longer term archives:

  • Annotation on the human NCBI36 assembly is available at our Ensembl 54 archive site.
  • Annotation on the mouse NCBIm37 assembly is available at our Ensembl 67 archive site.
  • As from August 2014 we are supporting the human GRCh37 assembly at our dedicated GRCh37 human site. Unlike the other Ensembl archive sites, this will be updated to the latest web interface every Ensembl release and there may be occasional data updates to human.
Archived databases are also maintained for at least 10 years. Currently all databases are available from 2004. More information is available from our MySQL database documentation. We also maintain data archives from 2004 available from our FTP site.

For all enquiries, please contact the Ensembl HelpDesk.


  • Ensembl aims to maintain stable identifiers for genes (ENSG), transcripts (ENST), proteins (ENSP) and exons (ENSE) as long as possible. Changes within the genome sequence assembly or an updated genome annotation may dramatically change a gene model. In these cases, the old set of stable IDs is retired and a new one assigned. Gene and transcript pages both have an ID History view which maps changes in the ID from the earliest version in Ensembl.
  • Protein family identifiers (fam), Ensembl EST gene identifiers (ENSESTG) and Genscan identifiers (GENSCAN) are currently not stable.
  • With the exception of the GRCh37 human site BLAST, BLAT and other tools are not available from the archive sites.
  • Accounts are shared between the current site and almost all archives. The exceptions are the older human NCBI36 and the mouse GRCm37 sites where changes in architecture and code make sharing logins impractical.

Linking to the Archive Ensembl sites

The Archive Ensembl sites have the format: http://<three-letter-month><year>.archive.ensembl.org for example http://nov2008.archive.ensembl.org)

In the footer of each current Ensembl page, there is a link called 'Permanent link', which links to the corresponding page in the Ensembl Archive. A similar link on each archive page links back to the current site (i.e. www.ensembl.org).

For example if you are looking at the Alternative Splicing view for human gene BRCA2 on the main Ensembl site in August 2015, when Ensembl 80 was the current version, the URL would be:


and the equivalent archived page URL would be:


Unfortunately, owing to the change in site organisation between releases it is not always possible to map pages one-to-one between the current Ensembl site and the older archives. If the link does not take you to the data you expected, trying using the search facility to locate the information.