Chr Position Gene Prediction Model Tree vote Type
AAE dbSNP Ref Alt Clinvar Features Region Coverage Hom / Het

MutationTaster Examples

All examples can also be accessed from the footer of the main page.

Deleterious variants

CHRND_L63P

This single base exchange in the CHRND gene is listed in NCBI ClinVar as a known disease-causing variant (dbSNP:rs121909508) for MYASTHENIC SYNDROME, CONGENITAL, FAST-CHANNEL, OMIM 100720#0013).

It results in an amino acid exchange from leucin to proline. The involved amino acid residue is part of a functional domain (topo domain) and is highly conserved (amino acids identical in all homologes). As a consequence of the mutation, the topo domain is lost. The phyloP and phastCons values for the changed nucleotide are very high, again indicating strong evolutionary conservation.

Show in MutationTaster: Analyse all transcripts, analyse transcript ENST00000258385

CLDN16_G191R

This variant is a single base exchange that is listed in NCBI ClinVar as a known disease-causing variant (dbSNP:rs104893721) for HOMG3 (HYPOMAGNESEMIA 3, OMIM 603959#0003).

The altered base leads to an amino acid substitution at a highly conserved residue (amino acid is identical in all homologs). Since the amino acid exchange is quite dramatic (a neutral glycine is replaced with a charged arginine), the substitution has a high amino acid exchange score (3.41, extracted from the Grantham matrix).

As reflected by the high phyloP and phastCons scores, the position is also conserved on the DNA level. The predicted splice site changes are only marginal and can therefore be neglected. The classification 'disease causing' is due to the ClinVar entry but the high probability indicates, that MutationTaster would have classified this variant as deleterious anyway.

Show in MutationTaster: Analyse transcript ENST00000264734

SOD1_H47R

This example is a mutation that is listed in NCBI ClinVar as a known disease-causing variant (dbSNP:rs121912443) for ALS (AMYOTROPHIC LATERAL SCLEROSIS 1, OMIM 147450#0008). The involved amino acid residue is highly conserved and lies within the catalytic copper-binding site of the protein, which is displayed as lost protein feature.

The phyloP and phastCons values for this position (4.283 and 0.996) are also very high, indicating strong evolutionary conservation. Again, the prediction is due to the ClinVar entry but has a very high probability anyway.

Show in MutationTaster: Analyse transcript ENST00000270142

Benign variants

AGRN_g.28670_28671delAG

This example is a deletion that was found more than 4 times in the 1000Genomes Project (1000G) in homozygous state. There is no amino acid exchange since the deletion lies within an intronic sequence. As it can be seen from the low phyloP / phastCons values, the deleted bases are hardly conserved. The splice site changes should be interpreted with care, especially in this context where the variant is not conserved and has been frequently found homozygous in the 1000G. It can be assumed that the splice site changes are false positive predictions. The classification 'benign' is due to the findings in the 1000G but the high probability indicates, that MutationTaster would have classified this variant as non-disease causing anyway.

Show in MutationTaster: Analyse transcript ENST00000379370

VCF batch querys

Tiny sample file

The VCF file TinyExample38.vcf only contains three lines to illustrate the file format:

3    190122694    .    G    A    116    .    .    GT:DP    0/1:154
1    984171    .    CAG    C    116    .    .    GT:DP    0/1:154
21    33036170    .    A    G    116    .    .    GT:DP    0/1:154

Show in MutationTaster

Sample exome (cystic fibrosis)

The zipped file SampleExome_CF38.vcf.gz contains a public VCF file with disease mutations in the CFTR gene spiked in. It is good practice to zip or gzip large VCF files before upload to MutationTaster. All variants must be in VCF format and refer to GRCh38 / hg38.

Show in MutationTaster

MutationTaster API

Automated analysis of chromosomal positions

If you want to integrate queries for chromosomal positions of our web interface in your personal NGS pipelines, this is possible in two ways depending on the output you prefer.

Text output

https://www.genecascade.org/MutationTaster2025/modperl/API.cgi?variants=2:232526664T>C
https://www.genecascade.org/MutationTaster2025/modperl/API.cgi?variants=21:33039603A>C,2:232526664T>C

If submitted via GET, the character ">" in the commands above is usually encoded as "%3E". That is,

https://www.genecascade.org/MutationTaster2025/modperl/API.cgi?variants=2:232526664T%3EC
https://www.genecascade.org/MutationTaster2025/modperl/API.cgi?variants=21:33039603A%3EC,2:232526664T%3EC

The value for variants can be one or multiple variants (separated by comma). The output is a table with 14 columns: chr, pos, ref, alt, transcript_stable, NCBI_geneid, prediction, model, tree_vote, note, splicesite, distance_from_splicesite, disease_mutation, polymorphism

You can specify your preferred reference genome and Ensembl build by adding &genome_version=38&ensembl_version=112 to your query. Currently, MutationTaster2025 only supports Ensembl build 112 of GRCh38, but support for GRCh37 and other Ensembl builds is planned for future releases.

In rare cases where the above URLs do not work, it may additionally be necessary to encode the characters ":" as "%3A" and "," as "%2C".

See output

Perl script

An example how to access the API via POST and print results can be found in this Perl script:

QueryMutationTasterAPI.pl

HTML output

https://genecascade.org/MutationTaster2025/cgi-bin/ChrPos.cgi?chromosome=2&position=232526664&ref=T&alt=C

The respective values for chromosome, position, ref and alt have to be set according to the variant(s) in question. The output has the same format as the output generated via the MutationTaster Web interface.

See output

Automated analysis of VCF files

POST request with curl

On Unix-based systems, you can use curl to post a VCF file to MutationTaster:

curl \
-F "name=Project_name" \
-F "email=your@email.edu" \
-F "filename=@Your_VCF_file.vcf" \
https://genecascade.org/MutationTaster2025/cgi-bin/MT_VCF_Pipeline.cgi

Replace Project_name, your@email.edu and Your_VCF_file.vcf with your details. The path to the VCF file must be prefixed with "@".

If the request is correct, you will immediately receive an email confirming your submission with a link to monitor the status of the analysis. Otherwise, please check the output of curl for possible errors.

Perl script

If you want to avoid the manual upload of VCF files to MutationTaster, we provide a Perl script which automatically sends a VCF file to the VCF analysis pipeline and afterwards retrieves the results. Please find it here:

sendVCF_MutationTaster.pl

 

 

If you encounter any problems regarding the API, please write us an email.

MutationTaster Support

Contact

In case you discover bugs, have suggestions or questions, please write an e-mail to Dominik Seelow (dominik.seelow AT charite.de).

We also appreciate hearing about your general experiences using MutationTaster.

Publications

Steinhaus R, Proft S, Schuelke M, Cooper DN, Schwarz JM, Seelow D. MutationTaster2021. Nucleic Acids Research. 2021 Apr 24.

Schwarz JM, Cooper DN, Schuelke M, Seelow D. MutationTaster2: mutation prediction for the deep-sequencing age. Nature methods. 2014 Apr;11(4):361-2.

Schwarz JM, Rödelsperger C, Schuelke M, Seelow D. MutationTaster evaluates disease-causing potential of sequence alterations. Nature methods. 2010 Aug;7(8):575-6.

Error messages

Message	Explanation
InsDel too long	At present, MutationTaster handles only InsDels up to 12 bases.
Your mutation of interest seems to span an exon/intron boundary.	This kind of mutation can only be analysed in gDNA mode.
No transcripts for this gene found!	You might have mis-spelled the gene symbol or used a protein name which is not always also the correct symbol (e.g. protein p53 is gene TP53). Also, in some (rare) cases a NCBI gene could not be mapped to an Ensembl gene. As some external data is based on NCBI while other is based on Ensembl, MT needs both to make a prediction. Moreover, we filter out protein-coding transcripts (Ensembl biotype protein_coding) without a correct start codon (ATG) and correct stop codon (TGA, TAA, TAG). This might lead to the phenomenon that MutationTaster complains about "no suitable transcripts" or "no transcripts for this gene found" although Ensembl lists one or several. Transcripts of mitochondrial genes are not tested for integrity due to differences in the mitochondrial genetic code.
No internal Ensembl transcript ID found. / No Ensembl gene ID found for transcript. / No stable ID for this gene.	Our database doesn't know the transcript you specified. This might happen if you refer to a newer or older release than the one we use. The release MT uses is mentioned on the query interface.
Ensembl gene XXX not found in ENSEMBL	Our database doesn't know the gene you specified. This might happen if you refer to a newer or older release than the one we use. The release MT uses is mentioned on the query interface.
No NCBI gene ID found. / No NCBI gene ID found for this transcript.	In some (rare) cases an Ensembl gene could not be mapped to a NCBI gene. As some external data is based on NCBI while other is based on Ensembl, MT needs both to make a prediction.
Too many NCBI gene IDs found.	In some (rare) cases an Ensembl gene could not be mapped to a single NCBI gene. As some external data is based on NCBI while other is based on Ensembl, MT needs both to make a prediction.
Only invalid NCBI gene IDs found.	In some (very rare) cases an Ensembl gene could not be mapped to a valid NCBI gene, i.e. the NCBI gene Ensembl refers to is 'discontinued' and was replaced by another gene. As some external data is based on NCBI while other is based on Ensembl, MT needs both to make a prediction. Please contact us if you encouter such a case.
Gene XXX not found on any chromosome.	The gene under scrutiny has no valid positional data. This should not occur at all. Please contact us if you encouter such a case.
Gene XXX (Entrez gene YYY) and transcript ZZZ do not match!	The transcript you entered is not a product of the gene you entered. Please check your input.
Position is out of gene!	You entered a position that is located outside the gene. This may happen when you mapped genomic position to gene-specific position using an old genome build. Or, of course, by typos. Please check your input.
Could not retrieve a sequence or sequence is too short.	MT was not able to get the gene sequence from Ensembl. This might be due to network problems so you should repeat the analysis after some time. Should this not work, please contact us.
No start ATG exon found.	The transcript is not properly annotated: there is no start position of the coding sequence in the database. Please select another transcript of the same gene.
No stop exon found.	The transcript is not properly annotated: there is no stop position of the coding sequence in the database. Please select another transcript of the same gene.
Chosen transcript ENSTXXX has no correct start ATG annotated.	Protein-coding transcripts (Ensembl biotype protein_coding) are tested for transcript integrity, i.e. for presence of a correct start codon (ATG) and correct stop codon (TGA, TAA, ATG). If one is missing, an error message is thrown out because analysis in corrupt transcripts might lead to a wrong prediction.
Sequence XXX is not unique in your gene!	Please use a longer snippet.
Sequence was not found in your gene.	Please check your input: is there a typo in your snippet? Or do you use a snippet created from the wrong strand? MT always refers to the strand the gene is located on.
Snippet not properly formatted.	Please check your input: snippets must be specified as ACGTACGT[OLDBASES/NEWBASES]ACGTACGT.

MutationTaster FAQs

General questions

Can I download pre-computed predictions?

Unfortunately not. Unlike SIFT or PolyPhen which handle only single amino acid substitutions, MutationTaster works on DNA level and allows insertions and deletions. The exome alone comprises about 30 Mb with 3 possible single base exchanges at each site (let alone introns and InDels). These 30 M x 3 SBEs may affect several different transcripts, leading to about 30,000,000 (Mb) x 3 (SBEs) x 5 (transcripts) = 450,000,000 values to pre-compute.

We could of course generate such a list, but it would still not include the InDels and most of the introns. What is more important: such a list would take a very long time to generate and might soon become outdated. We rather spend our efforts on improving MutationTaster!

Why doesn't MutationTaster know my valid transcript ID?

We filter out protein-coding transcripts (Ensembl biotype protein_coding) without a correct start codon (ATG) and correct stop codon (TGA, TAA, ATG). This might lead to the phenomenon that MutationTaster complains about "no suitable transcripts" or "no transcripts for this gene found" although there are some listed in Ensembl. We decided to exclude such transcripts from analysis in MutationTaster due to their bad annotation, which might in the end lead to a wrong prediction. Transcripts of mitochondrial genes are not tested for integrity due to differences in the mitochondrial genetic code.

Does a high Tree vote value indicate a high probability for a correct prediction, then?

Unfortunately not. Our results show that wrong predictions are usually not reflected by low Tree vote values but are rather caused by benign or deleterious variants that show characteristics of the other case, e.g. SNPs that are highly conserved and destroy protein features or disease mutations that appear to have no effect on the protein/gene at all.

Why don't you exclude known SNPs as possible deleterious mutations?

Because many variants listed in db SNP have never been shown to show all three genotypes in unaffected individuals. Some SNPs even appear to have only one allele. And even if both alleles were observed, there should be a sufficient number of healthy individuals who are homozygous for the minor allele to exclude a damaging effect.

If all three genotypes were observed in the HapMap project, or the variant was found homozygously in the 1000Genomes Project more than 4 times, it will automatically be regarded as benign.

Why don't you use allele frequencies to exclude variants as possible deleterious mutations?

We decided not to do this because not all loss-of-function variants are rare (e.g. CFTR mutations). Instead, we rely on healthy homozygous individuals. Are they observed, a variant is unlikely to cause a severe early-onset single gene disorder and hence predicted to be benign.

The prediction for my favourite variant has changed. Why?

Well, this is a very rare event. However, as the available data such as protein features is increasing, we regularly update our database and re-train the classifier. In some cases, the annotation of a gene improves drastically. This may yield formerly unknown protein features in your gene/protein at your position which can of course influence the prediction of your variant.

Is there any way to learn how a single variant is classified?

Well, yes and no. A Random Forest classifier studies the frequencies of single item statuses (such as 'conservation in cattle - highly conserved' or 'existence of a disulfide bond - no') in both groups of the training set ('polymorphisms' / 'disease mutations'). It compares the statuses of these items in your variant with the known frequencies and then decides which group fits best.

You can of course study the model and hence the frequencies used by the classifier. They can be found in our supplementary data.

What does "InDel variants are limited to 40 bp" mean?

The website states that with MT "InDel variants are limited to 40 bp", however, does that mean the ACTUAL insertion or deletion, or the DESCRIPTION of the insertion or deletion? For example, the description of the ALT variant contains 15 bases, whilst the REF variant contains 13 bases, so that the actual INSERTION (AC) here is only 2 bases but it has been rejected for being too long.

When bases get inserted to / deleted from a stretch of similar bases, as in the given example to a stretch of several 'AC', MutationTaster doesn't know at which position exactly the 'AC' was inserted (or deleted), due to the whole stretch of AC. That's why it has to use the whole 13 bases, although the actual insertion is only 2 bases. This is also the reason why such variants are described that way in your VCF file.

Why don't you offer a MutationTaster download version for local installation on my own machine?

We are asked regularly for standalone versions of MutationTaster, our conversion tools or the database. We don't offer these services, because it is not feasible. We would flood the world with lots of different versions of MutationTaster which we could never maintain. The distribution of local installations probably would lead to hundreds of support questions and we (only 2 people) are already busy with those that concern the version we control and know. We are not able to give support concerning installation issues or questions like 'how is the conservation internally stored?' or errors that occur only in the versions modified by the users. Moreover, you would need a very powerful hardware and highly optimised server to reach the same speed as the online version. MutationTaster uses a database which is tens of GBs in size with parts of Ensembl and the 1000 Genomes data in it. Additionally, we use some external tools for which we have signed disclosure agreements and which we are hence not allowed to share with other groups anyway. If you want to integrate MutationTaster in your own analysis pipeline for Next Generation Sequencing data, we suggest to use our VCF analysis pipeline that can be called via Perl's WWW::Mechanize module and similar approaches.

What does the AA changes score mean and how does it influence the prediction?

The score is taken from the Grantham Matrix for amino acid substitutions and reflects the physicochemical difference between the original and the mutated amino acid. It ranges from 0.0 to 215 but does not provide a value for amino acid insertions/deletions. However, the score is only displayed for information purpose and does not influence the prediction. Instead, MutationTaster uses the frequency of the respective AA exchange in known disease causing mutations and polymorphisms for the classification.

Why is the same variant classified as benign when there is an amino acid exchange and as deleterious when there is no amino acid exchange?

MutationTaster uses five different models (without_aae, simple_aae, complex_aae, 3utr, 5utr) for its prediction. Depending on the type of variant, MutationTaster automatically determines the correct model. Each model was trained with a suitable set of known polymorphisms/ disease mutations and the prioritisation of the individual parameters differs among the different models. Thus the prediction of a variant might not be the same, if two different models are applied (e.g. without_aae model and simple_aae model). In some cases with a 'deleterious' prediction due to DNA related features such as strong conservation, knowledge of the effect of amino acid substitution can 'weaken' the prediction, e.g. if the difference of the two amino acids is modest and no protein domains are affected. This is a consequence of the different models: If we used only one, all 'silent' mutations would be considered as benign - and we decided to rather risk false positives than to lose any true positives.

Why are the phlyoP and phastCons scores for the same variant different in MutationTaster2025 compared to MutationTaster2021 or other sorces like gnomAD?

There are multiple different conservation tracks available in UCSC which use a different amount of species alignments. In MutationTaster2025 we use the 100 vertebrate species version for GrCh38, whereas in MutationTaster2021 we used the 46 track for GrCh37. Similarly, other data sources may use different alignments (e.g. gnomAD uses the 241 way alignment) and their scores may differ.

The AlphaMissense transcript links sometimes don't show a result, why?

There are transcripts, for which there simply is no structure available in AlphaMissense. This is why MutationTaster2025 always provides the gene-specific link to AlphaMissense along with the transcript-specific one.

VCF files

Why are there so many cases with no prediction (n/a)?

Most of the n/a cases are due to a missing link between an Ensembl transcript and an NCBI gene (error message: no NCBI gene ID found for this transcript). Ensembl has far more genes and transcripts annotated than NCBI, however, we need to link the Ensembl genes to NCBI in order to get the HGNC genesymbol and SwissProt Accession ID. To circumvent this, we plan to fetch SwissProt ID and genesymbol also via Ensembl in the future, so that in case of a missing link to NCBI, the analysis can be conducted neverthess.

Why are there so many outsides genes, although I have uploaded a VCF file from Exome Sequencing?

Target enrichment is not 100% perfect, thus it is normal that there are variants outside genes. Moreover, we do not use all available transcripts (see MutationTaster FAQs), because some are not suitable for analysis with MutationTaster. In case there is a variant in a gene which only has transcripts not suitable for MutationTaster analysis, this will be counted as outside gene

MutationTaster Changelog

Welcome to the MutationTaster changelog. This page informs you about the latest developments and modifications to MutationTaster.

20 December 2024

MutationTaster2025

Our latest release of MutationTaster includes the following changes:

• Genome version GRCh38
• Genotype counts from gnomAD 4.1
• mtDNA variant counts and heteroplasmy from gnomAD 3.1
• transcript-specific protein domains and protein conservation
• multi-sample VCF support
• gene-specific mutation tolerance (gnomAD gene constraints) included in prediction
• New URL: https://www.genecascade.org/MutationTaster2025/

Previous updates

Visit https://www.genecascade.org/MutationTaster2021/info/#changelog

MutationTaster Legal

Imprint

Responsible for the purposes of media law for this page

Dominik Seelow
Bioinformatics and Translational Genetics
Berliner Institut für Gesundheitsforschung
Charitéplatz 1
10117 Berlin
email: dominik.seelow (at) charite.de

Data protection

If you use one of our query engines, you may enter an e-mail address. The secret URL pointing at your project will be sent to you directly after the upload and you will receive a second notification when the analysis has finished.
Your e-mail is only visible during the data analysis and the analysis web pages can only be accessed via the secret URL.
Data you uploaded will be automatically deleted within two months time unless you delete it by yourself or request complete deletion or an extension of that period by e-mail. Please note that your from-address must match the address you specified when uploading the data. Your e-mail address will remain in our database to allow us to determine the number of different users. It will be deleted upon request.

License

MutationTaster 2025 is free and open to all users and there is no login requirement. Please contact us if you want to include pre-computed scores of MutationTaster into your own software.