Ensembl VEP

Haplosaurus (haplo) is a tool that implements the functionality of the Ensembl transcript haplotypes view. Unlike VEP, which analyzes variants independently, haplo considers the combined effect of all variant alleles across a transcript by constructing haplotype sequences.

Key capabilities:

• Takes phased genotypes from a VCF and constructs pairs of haplotype sequences (CDS and protein).
• Aligns and compares variant haplotype sequences to the reference.
• Constructs HGVS-like names representing differences to the reference.
• Uses VEP caches, Ensembl databases, GFF, and GTF files as transcript data sources.

Requirements:

• Input must be a VCF containing phased genotype data.
• The VCF file must be sorted by chromosome and genomic position.

./haplo -i input.vcf -o out.txt -cache

Execute the pipeline using the nextflow run main.nf command. You must provide an input file or directory and a VEP configuration file.

nextflow run main.nf \
  -profile <profiles-to-use> \
  --input <path-to-file> \
  --vep_config <path-to-vep-config>

Docker

A Docker image is available on DockerHub at ensemblorg/ensembl-vep. Refer to the official documentation for specific installation instructions.

Conda

An Ensembl VEP Conda package exists, but it is not produced or maintained by the Ensembl VEP team. Use it at your own risk, as the team cannot provide support for issues related to the Conda installation.

The variant_recoder tool translates between different variant encodings (VCF, variant ID, HGVS). It supports genomic, transcript, and protein HGVS, as well as RefSeq and LRG sequences.

Important Notes:

• It requires database access for identifier lookup and cannot be used in offline mode.
• It requires the JSON Perl module.
• Output is in JSON format.

Usage Examples:

• For a single variant string: ./variant_recoder --id [input_data_string]
• For an input file (one variant per line): ./variant_recoder -i [input_file] --species [species]

./variant_recoder --id "AGT:p.Met259Thr" --pretty
./variant_recoder -i [input_file] --species [species]

While the core installation handles basic requirements, several optional CPAN modules provide significant functionality or performance improvements. It is recommended to use cpanm to install them:

• DBD::mysql: Required for database access (using --database or --cache without --offline).
• Set::IntervalTree: Required for Haplosaurus and improves VEP speed.
• JSON: Required for writing JSON output.
• PerlIO::gzip: Enables faster compressed file parsing.
• Bio::DB::BigFile: Required for reading custom annotation data from BigWig files.

# Example using cpanm (recommended)
cpanm DBD::mysql Set::IntervalTree JSON PerlIO::gzip Bio::DB::BigFile

To speed up haplo analysis (especially for long transcripts with indels), you can install the bioperl-ext package. Note that this package is no longer maintained and requires manual installation. The following steps install it to $HOME/perl5.

Warning: You may need to adjust the export PERL5LIB command to match your specific architecture and Perl version.

git clone https://github.com/bioperl/bioperl-ext.git
cd bioperl-ext/Bio/Ext/Align/
perl -pi -e"s|(cd libs.+)CFLAGS=\\'|\$1CFLAGS=\\'-fPIC |" Makefile.PL
perl Makefile.PL PREFIX=~/perl5
make
make install
cd -
export PERL5LIB=${PERL5LIB}:${HOME}/perl5/lib/x86_64-linux-gnu/perl/5.22.1/

To install the ensembl-vep package, you need gcc, g++, make, and Perl (>=5.22 recommended). You must also have the Perl libraries Archive::Zip and DBI installed.

Follow these steps to install from source:

1. Clone the repository.
2. Run the INSTALL.pl script.

You can re-run INSTALL.pl to check for updates to the package and its dependencies.

git clone https://github.com/Ensembl/ensembl-vep.git
cd ensembl-vep
perl INSTALL.pl

To run database tests, you must have write access to a MySQL server. You need to configure the test database parameters by creating a configuration file from the provided example.

1. Copy MultiTestDB.conf.example to MultiTestDB.conf.
2. Edit MultiTestDB.conf to set the required connection and database parameters.

cp MultiTestDB.conf.example MultiTestDB.conf

You can generate test databases by cloning a core database from the Ensembl server using the clone_core_database.pl script. This requires a JSON configuration file specifying the regions to be used.

Note on species naming: To prevent collisions with existing databases on your host, the script is configured to rename the species to homo_vepiens. After cloning, you must rename the directory in t/test-genome-dbs and update the metadata in t/test-genome-dbs/homo_vepiens/core/meta.txt to reflect the changes.

perl ../../ensembl-test/scripts/clone_core_database.pl -host ensembldb.ensembl.org \\n-user anonymous -dbname homo_sapiens_core_84_38 -json test_core_region_config.json \\n-directory t/test-genome-dbs -dest_host [host] -dest_user [user] -dest_pass [pass]

You can validate the merged VCF output using bcftools. The following command checks the first variant in the generated output directory.

  work/singularity/quay.io-biocontainers-bcftools-*.img bcftools view \
    -H outdir/merged-file.vcf.gz \
    -r 1