Ensembl mobile site help

Things to know when navigating the Ensembl mobile site

Search box

Use the search box at the top right of all Ensembl views to search for a gene, phenotype, sequence variant, and more.

Top navigation

Touch MENU button to open the main menu and touch again to close.

Touch MENU

Left hand side menu

Touch the left menu icon () or swipe right to open the side menu and touch anywhere outside the menu or touch the cross icon or swipe left to close.

The ? icon

Touch the icon to get help

And don't forget to send us your comments using the feedback link inside the main menu.

EnsemblEnsembl Home

Horse assembly and gene annotation

The horse (Equus caballus) is an odd-toed ungulate mammal belonging to the taxonomic family Equidae. It has evolved over the past 45 to 55 million years from a small multi-toed creature, Eohippus, into the large, single-toed animal of today.

Humans began domesticating horses around 4000 BC, and their domestication is believed to have been widespread by 3000 BC. So-called "wild horses" are usually feral populations descended from domesticated E. caballus; true wild horses such as Przewalski's horse and the now-extinct tarpan are generally considered to be separate Equus species.


The EquCab3.0 assembly was submitted by University of Louisville on January 2018. The assembly is on chromosome level, consisting of 10,987 contigs assembled into 4,701 scaffolds. From these sequences, 32 chromosomes have been built. The N50 size is the length such that 50% of the assembled genome lies in blocks of the N50 size or longer. The N50 length for the contigs is 1,502,753 while the scaffold N50 is 87,230,776.

Gene annotation

The gene annotation process was carried out using a combination of protein-to-genome alignments, annotation mapping from a suitable reference species and RNA-seq alignments (where RNA-seq data with appropriate meta data were publicly available). For each candidate gene region, a selection process was applied to choose the most appropriate set of transcripts based on evolutionary distance, experimental evidence for the source data and quality of the alignments.
Small ncRNAs were obtained using a combination of BLAST and Infernal/RNAfold.
Pseudogenes were calculated by looking at genes with a large percentage of non-biological introns (introns of <10bp), where the gene was covered in repeats, or where the gene was single exon and evidence of a functional multi-exon paralog was found elsewhere in the genome.
lincRNAs were generated via RNA-seq data where no evidence of protein homology or protein domains could be found in the transcript.

In accordance with the Fort Lauderdale Agreement, please check the publication status of the genome/assembly before publishing any genome-wide analyses using these data.

More information

General information about this species can be found in Wikipedia.



AssemblyEquCab3.0, INSDC Assembly GCA_002863925.1, Jan 2018
Base Pairs2,506,966,135
Golden Path Length2,506,966,135
Annotation providerEnsembl
Annotation methodFull genebuild
Genebuild startedDec 2018
Genebuild releasedDec 2018
Genebuild last updated/patchedSep 2019
Database version104.3

Gene counts

Coding genes20,955
Non coding genes9,014
Small non coding genes1,761
Long non coding genes7,244
Misc non coding genes9
Gene transcripts59,087


Genscan gene predictions49,384
Short Variants21,198,236
Structural variants193,747

About this species