EnsemblEnsembl Home

Ensembl Variation - Predicted data

Ensembl imports variant data from a variety of different sources, as described on the Data description page. Below we give more information about how Ensembl predicts the effects of variants. Ensembl calculates the:

Variation consequences and predictions of a variant, by gene and transcript
  • Variants consequences in transcripts (synonymous, missense,...), e.g. rs699
  • Protein function prediction (SIFT, PolyPhen), e.g. rs56404215
Linkage disequilibrium
  • Linkage disequilibrium information, e.g. rs1333049
  • Tagged variants

Calculated variant consequences

For each variant that is mapped to the reference genome, we identify each Ensembl transcript that overlap the variant. We then use a rule-based approach to predict the effects that each allele of the variant may have on the transcript. The set of consequence terms, defined by the Sequence Ontology (SO), that can be currently assigned to each combination of an allele and a transcript is shown in the table below. Note that each allele of each variant may have a different effect in different transcripts.

This approach is applied to all germline variants and somatic mutations stored in the Ensembl variation databases (though we do not yet currently calculate consequences for structural variants). The resulting consequence type calls, along with information determined as part of the process, such as the cDNA and CDS coordinates, and the affected codons and amino acids in coding transcripts, are stored in the Ensembl Variation database and displayed on the website. You can use this pipeline for your own data via the VEP.

We used SO terms by default since the Ensembl release 68. There is an equivalent SO term for each of our old Ensembl terms but in a few cases there is a more specific SO term available, as shown in the table below. If you have text files or VEP outputs with our old Ensembl terms, you can easily update these to using the SO terms by running the following script e.g.

perl convert_ensembl_to_SO_consequences.pl input.txt > converted.txt

See below a diagram showing the location of each display term relative to the transcript structure:
consequence diagram

The terms in the table below are shown in order of severity (more severe to less severe) as estimated by Ensembl, and this ordering is used on the website summary views. This ordering is necessarily subjective and API and VEP users can always get the full set of consequences for each allele and make their own severity judgement. The IMPACT rating is a separate rating given for compatibility with other variant annotation tools (e.g. snpEff).

* SO term SO description SO accession Display term IMPACT
transcript_ablation A feature ablation whereby the deleted region includes a transcript feature SO:0001893 Transcript ablation HIGH
splice_acceptor_variant A splice variant that changes the 2 base region at the 3' end of an intron SO:0001574 Splice acceptor variant HIGH
splice_donor_variant A splice variant that changes the 2 base region at the 5' end of an intron SO:0001575 Splice donor variant HIGH
stop_gained A sequence variant whereby at least one base of a codon is changed, resulting in a premature stop codon, leading to a shortened transcript SO:0001587 Stop gained HIGH
frameshift_variant A sequence variant which causes a disruption of the translational reading frame, because the number of nucleotides inserted or deleted is not a multiple of three SO:0001589 Frameshift variant HIGH
stop_lost A sequence variant where at least one base of the terminator codon (stop) is changed, resulting in an elongated transcript SO:0001578 Stop lost HIGH
start_lost A codon variant that changes at least one base of the canonical start codon SO:0002012 Start lost LOW
transcript_amplification A feature amplification of a region containing a transcript SO:0001889 Transcript amplification HIGH
inframe_insertion An inframe non synonymous variant that inserts bases into in the coding sequence SO:0001821 Inframe insertion MODERATE
inframe_deletion An inframe non synonymous variant that deletes bases from the coding sequence SO:0001822 Inframe deletion MODERATE
missense_variant A sequence variant, that changes one or more bases, resulting in a different amino acid sequence but where the length is preserved SO:0001583 Missense variant MODERATE
protein_altering_variant A sequence_variant which is predicted to change the protein encoded in the coding sequence SO:0001818 protein altering variant MODERATE
splice_region_variant A sequence variant in which a change has occurred within the region of the splice site, either within 1-3 bases of the exon or 3-8 bases of the intron SO:0001630 Splice region variant LOW
incomplete_terminal_codon_variant A sequence variant where at least one base of the final codon of an incompletely annotated transcript is changed SO:0001626 Incomplete terminal codon variant LOW
stop_retained_variant A sequence variant where at least one base in the terminator codon is changed, but the terminator remains SO:0001567 Stop retained variant LOW
synonymous_variant A sequence variant where there is no resulting change to the encoded amino acid SO:0001819 Synonymous variant LOW
coding_sequence_variant A sequence variant that changes the coding sequence SO:0001580 Coding sequence variant MODIFIER
mature_miRNA_variant A transcript variant located with the sequence of the mature miRNA SO:0001620 Mature miRNA variant MODIFIER
5_prime_UTR_variant A UTR variant of the 5' UTR SO:0001623 5 prime UTR variant MODIFIER
3_prime_UTR_variant A UTR variant of the 3' UTR SO:0001624 3 prime UTR variant MODIFIER
non_coding_transcript_exon_variant A sequence variant that changes non-coding exon sequence in a non-coding transcript SO:0001792 Non coding transcript exon variant MODIFIER
intron_variant A transcript variant occurring within an intron SO:0001627 Intron variant MODIFIER
NMD_transcript_variant A variant in a transcript that is the target of NMD SO:0001621 NMD transcript variant MODIFIER
non_coding_transcript_variant A transcript variant of a non coding RNA gene SO:0001619 Non coding transcript variant MODIFIER
upstream_gene_variant A sequence variant located 5' of a gene SO:0001631 Upstream gene variant MODIFIER
downstream_gene_variant A sequence variant located 3' of a gene SO:0001632 Downstream gene variant MODIFIER
TFBS_ablation A feature ablation whereby the deleted region includes a transcription factor binding site SO:0001895 TFBS ablation MODERATE
TFBS_amplification A feature amplification of a region containing a transcription factor binding site SO:0001892 TFBS amplification MODIFIER
TF_binding_site_variant A sequence variant located within a transcription factor binding site SO:0001782 TF binding site MODIFIER
regulatory_region_ablation A feature ablation whereby the deleted region includes a regulatory region SO:0001894 Regulatory region ablation MODERATE
regulatory_region_amplification A feature amplification of a region containing a regulatory region SO:0001891 Regulatory region amplification MODIFIER
feature_elongation A sequence variant that causes the extension of a genomic feature, with regard to the reference sequence SO:0001907 Feature elongation MODIFIER
regulatory_region_variant A sequence variant located within a regulatory region SO:0001566 Regulatory region variant MODIFIER
feature_truncation A sequence variant that causes the reduction of a genomic feature, with regard to the reference sequence SO:0001906 Feature truncation MODIFIER
intergenic_variant A sequence variant located in the intergenic region, between genes SO:0001628 Intergenic variant MODIFIER

* Corresponding colours for the Ensembl web displays.

Protein function predictions

For human mutations that are predicted to result in an amino acid substitution we provide SIFT and PolyPhen predictions for the effect of this substitution on protein function. We compute the predictions for each of these tools for all possible single amino acid substitutions in the Ensembl human proteome. This means we can provide predictions for novel mutations for VEP and API users. We were able to compute predictions from at least one tool for over 95% of the human proteins in Ensembl. SIFT predictions are also available for chicken, cow, dog, horse, mouse, pig, rat sheep and zebrafish.

These tools are developed by external groups and we provide a brief explanation of the approach each takes below, and how we run it in Ensembl. For much more detail please see the representative papers listed below, and the relevant publications available on each tool's website.


SIFT predicts whether an amino acid substitution is likely to affect protein function based on sequence homology and the physico-chemical similarity between the alternate amino acids. The data we provide for each amino acid substitution is a score and a qualitative prediction (either 'tolerated' or 'deleterious'). The score is the normalized probability that the amino acid change is tolerated so scores nearer 0 are more likely to be deleterious. The qualitative prediction is derived from this score such that substitutions with a score < 0.05 are called 'deleterious' and all others are called 'tolerated'.

We ran SIFT version 5.2.2 following the instructions from the authors and used SIFT to choose homologous proteins rather than supplying them ourselves. We used all protein sequences available from UniRef90 (release 2014_11) as the protein database.

List of species with SIFT data:
  • Bos taurus
  • Canis familiaris
  • Danio rerio
  • Equus caballus
  • Gallus gallus
  • Homo sapiens
  • Mus musculus
  • Ovis aries
  • Rattus norvegicus
  • Sus scrofa


PolyPhen-2 predicts the effect of an amino acid substitution on the structure and function of a protein using sequence homology, Pfam annotations, 3D structures from PDB where available, and a number of other databases and tools (including DSSP, ncoils etc.). As with SIFT, for each amino acid substitution where we have been able to calculate a prediction, we provide both a qualitative prediction (one of 'probably damaging', 'possibly damaging', 'benign' or 'unknown') and a score. The PolyPhen score represents the probability that a substitution is damaging, so values nearer 1 are more confidently predicted to be deleterious (note that this the opposite to SIFT). The qualitative prediction is based on the False Positive Rate of the classifier model used to make the predictions.

We ran PolyPhen-2 version 2.2.2, release 405 (available here) following all instructions from the authors and using the UniProtKB UniRef100 (release 2013_10) non-redundant protein set as the protein database and DSSP (snapshot 22-Nov-2013) and PDB (snapshot 22-Nov-2013) as the structural databases. When computing the predictions we store results for the classifier models trained on the HumDiv and HumVar datasets. Both result sets are available through the variation API which defaults to HumVar if no selection is made. (Please refer to the PolyPhen website or publications for more details of the classification system).

List of species with PolyPhen data:
  • Homo sapiens

Inconsistent predictions

PolyPhen and SIFT results are heavily dependant on sequence conservation estimates derived from protein sequence alignments and using different versions of the protein databases can result in substantial variance in the predictions and scores obtained. Ensembl releases are not synchronised with updates to the SIFT and PolyPhen web servers, so differences are to be expected. Changes in calls will also be seen between some Ensembl releases due to updates in the software and databases used.

Prediction data format

The SIFT and PolyPhen predictions are precomputed and stored in the variation databases and predictions are accessible in the variation API by using the sift_prediction, sift_score, polyphen_prediction and polyphen_score methods on a Bio::EnsEMBL::Variation::TranscriptVariationAllele object. For users wanting to access the complete set of predictions from the MySQL database or API, an explanation of the format used is provided here.

The predictions and associated scores are stored as a matrix, with a column for each possible alternate amino acid and a row for each position in the translation. Each prediction for a position and amino acid is stored as a 2-byte value which encodes both the qualitative prediction and score encoded as described below. A simple example matrix is shown in the figure below, though note we only show the decoded score rather than the actual binary value stored in the database.

protein function encoding

Prediction matrices can be fetched and manipulated in a user-friendly manner using the variation API, specifically using the ProteinFunctionPredictionMatrixAdaptor which allows you to fetch a prediction matrix using either a transcript or a translation stable ID. This adaptor returns a ProteinFunctionPredictionMatrix object and you can use the get_prediction method to retrieve a prediction for a given position and amino acid. Users who want to use entire matrices should use the deserialize method to decode an entire binary formatted matrix into a simple Perl hash. Please refer to the API documentation for both of these classes for more details. For users who require direct access to the MySQL database (for instance because they are accessing the database in a language other than Perl) we provide a description of the binary format used below.

Prediction matrices for each translation from each of SIFT and PolyPhen are stored in the protein_function_predictions table. The analysis used to calculate the predictions is identified in the analysis_attrib_id column which refers to a value stored in the attrib table, and the protein sequence the predictions apply to is identified by the translation_md5_id column which refers to a hexadecimal MD5 hash of the sequence stored in the translation_md5 table. The prediction matrices are stored in the prediction_matrix column as a gzip compressed binary string. Once uncompressed, this string contains a 40 byte substring for each row in the matrix concatenated together in position order. Each row is composed of 20 2-byte predictions, one for each possible alternative amino acid in alphabetical order, though note that the value for the amino acid that matches the reference amino acid is identified as a 2-byte value with all bits set, or 0xFFFF in hexadecimal notation. The prediction itself is packed as a 16 bit little-endian ("VAX" order, or "v" format if using the perl pack subroutine) unsigned short value. The top 2 bits of this short are used to encode the qualitative prediction (PolyPhen has 4 possible values, while SIFT has just 2), and the bottom 10 bits are used to encode the prediction score. To decode the qualitative prediction you should mask off all bits except the top 2, and shift the resulting short right by 14 bits and treat this as an integer between 0 and 3. The corresponding prediction can then be looked up in the table below. To decode the prediction score you should mask off the top 6 bits and the resulting value can be treated as a number between 0 and 1000, which should be divided by 1000 to give a 3 decimal place score (casting to a floating point type if necessary). Bits 11-14 are not used, except to encode the "same as reference" dummy prediction 0xFFFF.

protein function encoding

A diagram of the encoding scheme is shown above. In this example from a polyphen prediction, the top prediction bits are 0b01 which in decimal is the number 1, which corresponds to "possibly damaging" in the table below. The score bits are 0b1110001010 which in decimal is the number 906, which when divided by 1000, gives a score of 0.906.

Tool Numerical value Qualitative prediction
PolyPhen 0 "probably damaging"
PolyPhen 1 "possibly damaging"
PolyPhen 2 "benign"
PolyPhen 3 "unknown"
SIFT 0 "tolerated"
SIFT 1 "deleterious"

To retrieve a prediction for a particular amino acid substitution at a given position in a translation, first you must retrieve the binary matrix from the database and uncompress it using gzip. You can calculate the offset into this string by multiplying the desired position (starting at 0) by 20 and then adding the index of the desired amino acid in an alphabetical ordering of amino acids (also starting at 0), and then multiply this value by 2 to take into account the fact that each prediction uses 2 bytes. Each matrix also includes a 3 byte header used check if the data is corrupted etc. so you will also need to add 3 to the calculated offset. The 2 bytes at the calculated position can then be extracted and the approach described above can be used to retrieve the prediction and score. A perl implementation of this scheme can be found in the Bio::EnsEMBL::Variation::ProteinFunctionPredictionMatrix module in the variation API and should hopefully inform attempts to reimplement this scheme in other languages.


  • Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P, Kondrashov AS, Sunyaev SR.
    A method and server for predicting damaging missense mutations
    Nature Methods 7(4):248-249 (2010)

  • Kumar P, Henikoff S, Ng PC.
    Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm
    Nature Protocols 4(8):1073-1081 (2009)

  • Gonzalez-Perez A, Lopez-Bigas N.
    Improving the assessment of the outcome of non-synonymous SNVs with a Consensus deleteriousness score (Condel)
    Am J Hum Genet 88(4):440-449 (2011)