Both Ensembl release 93 and Ensembl Genomes release 40 are scheduled for late June and early July 2018, respectively.

Included are a number of new genomes and genebuilds for vertebrates and plants (including leopard, Amur tiger, hagfish, pigeon pea, carrot and adzuki bean) and significant updates to the mouse GENCODE annotation and regulatory build. This release will also bring a new import of variants from dbSNP for human, and allele frequencies for dog variation data!

Continue reading

Ensembl 91 is scheduled for December 2017 and we’re continuing our push to include the genome annotation for lots of new species. This time, we’re adding a whole new set of primate species to Ensembl.

Here’s what you can look forward to:

New assemblies, gene sets and annotations

  • Annotation of 12 new primate genomes, as well as updates to 6 existing genomes:
    • Nancy Ma’s night monkey
    • White-headed capuchin
    • Sooty mangabey
    • Angola colobus
    • Crab eating macaque
    • Southern pig-tailed macaque
    • Drill
    • Bonobo
    • Coquerel’s sifaka
    • Black snub-nosed monkey
    • Golden snub-nosed monkey
    • Black-capped squirrel monkey
    • Chimpanzee (update)
    • Gibbon (update)
    • Gorilla (update)
    • Mouse lemur (update)
    • Olive baboon (update)
    • Tarsier (update)
  • Annotation on the latest Cat genome assembly, Felis_catus_8.0
  • C. elegans gene set and annotation updated to Wormbase release WS260
  • Fruitfly gene set and annotation updated to Flybase release FB2017_04 (dmel_r6.17)
  • Updated Human cDNA alignments
  • Updated Mouse cDNA alignments
  • Updated microarray probe mappings and comparative genomics analyses for all new and updated species

Other updates and highlights

  • Updating our human variation database with:
    • COSMIC 82 somatic variants
    • HGMD 2017.2
    • DGVa structural variants
    • Phenotypes from NHGRI-EBI GWAS, OMIM, ClinVar, UniProt, Cosmic Gene Census, DDG2P, MIM Morbid and Orphanet
  • In other species we also have variation updates as follows:
    • dbSNP 150 in macaque, mouse, zebrafish, sheep, pig, horse, cow and chicken
    • DGVa in cow, dog and mouse, horse, macaque, pig, sheep and zebrafish
    • Phenotype updates from relevant databases in rat, zebrafish and mouse
  • Links to PharmGKB added from human variants
  • New web tool for Linkage Disequilibrium (LD) calculation
  • Updated GRCh37 regulatory features

For more details on the declared intentions, please visit our Ensembl admin site. Please note that these are intentions and are not guaranteed to make it into the release.

The upcoming Ensembl release (e!91) will include several updates to the regulation API and with it a farewell to many objects that have given the regulation API its characteristic look and feel over the years.

The changes listed below only affect the way we store high-throughput sequencing experiments and their results. Probe feature related objects and regulatory features are not affected. If you use any of the following in your scripts, please keep an eye for our updated doxygen documentation once the Ensembl release 91 is out.

ResultSet and InputSubset

The long serving ResultSet object and its faithful companion, the InputSubset object, will be removed. Over the last releases these data types have been extensively modified and moved to more specific API objects, until they only served to store information about the read files (InputSubset) and their respective alignments (ResultSet).

From now on alignments are handled by a new API object, called “Alignment”.

The InputSubset object will be replaced by two new objects:

  1. ReadFile
  2. ReadFileExperimentalConfiguration

A ReadFile represents a FASTQ file generated by a high-throughput sequencing experiment, such as ChIP-seq or DNAse-seq.

The experimental configuration that led to the creation of the read file is stored in the ReadFileExperimentalConfiguration object. It links the Experiment object to the ReadFiles generated by it and contains the following information:

  • which biological and
  • which technical replicate a ReadFile is within an Experiment,
  • whether it is paired-end, and
  • whether it is the result of multiple sequencing runs of the same sample.

Using the experimental configuration the Ensembl Regulation Sequence Alignment (ERSA) pipeline decides how to analyse the various high-throughput sequencing data.

AnnotatedFeature and FeatureSet

In the current API the AnnotatedFeature object represents enriched regions or peaks from ChIP-seq and DNase-seq experiments.

In the future the AnnotatedFeature API object will become the Peak object.

AnnotatedFeature objects used to be accessed by first fetching an appropriate FeatureSet object and then the AnnotatedFeatures linked to it.

A FeatureSet object that links to a set of AnnotatedFeatures represented a peak calling analysis from a ChIP-seq-like experiment. These are now represented by the new PeakCalling object.


The venerable DataSet object will be retired and it will not be replaced.


The CoordSystem object in regulation, not to be confused with the CoordSystem object used for Ensembl core databases, has been retired after many years of service.

It was mostly known for its Adaptor, which gave scary error messages, if the Registry had been misconfigured. It could also make features unexpectedly vanish from the website.

There are no plans to replace its function.


Current Object New Object Notes
InputSubset ReadFile
ResultSet Alignment
AnnotatedFeature Peak
FeatureSet PeakCalling
DataSet Retired.
CoordSystem Retired. Regulation-specific object. Not to be confused with that used for the Ensembl core databases.