The Ensembl Variant Effect Predictor (VEP) is a great way to analyse your own variation data. Available for free as a simple web tool, and as a downloadable perl script, the VEP allows you to input a list of variants and determine what genes they fall within and how they affect them.
The VEP allows you to input variants in a variety of formats, including a list of variant IDs, VCF files, HGVS notation and plain coordinates. This means it can be used to analyse data from many different sources, such as variant calling software or lists from publications.
Every variant you input will be plotted onto the genome and compared to the Ensembl gene set (in the case of human and mouse, this is the GENCODE set). The VEP will report back which transcripts your variants fall within or near, and the consequence of those variants on the transcripts, using Sequence Ontology terms. It will also report where in the gene and transcripts (and peptides where relevant) the variants fall. Where variants are missense, you will also receive information on the codon and amino acid change, and can choose to receive information on the severity of the amino acid change as SIFT and Polyphen scores. The VEP will also report if there is already a variant known at that location and can report allele frequencies from populations, such as 1000 genomes and NHLBI-ESP.
Many options are available when using the VEP. You can choose to see the consequences of your variants on regulatory regions, even filtering by regulatory activity in particular cell types. Where the variants already appear in our databases, you can find their overall and population-specific minor allele frequencies (where available) and filter out variants with a frequency above or below your chosen threshold. You can also filter by SO terms, or only see the most severe consequence of a variant.
The online tool is great for small numbers of variants (<750) and allows you to get your data in text or HTML form. If you have more variants than this, you should use the standalone perl script. You can carry out your analyses over the web or you can speed things up by doing it locally. Download one of our pre-built species-specific caches or, if we don’t have your species of interest, use our tools to create your own cache from a GTF file. You can even add your own genome annotation.
Developers looking to expand the VEP functionality can write their own plugins to work with the script. Many plugins are already available at our e! code site, where you can also submit your own plugins to help other users.
There is extensive documentation on using the VEP making it easy to manipulate to get the kinds of data you’re looking for. Give it a go.
Great post!