Skip to content

cancervariants/variation-normalization

Repository files navigation

Variation Normalization

DOI

Services and guidelines for normalizing variation terms to VRS compatible representations.

Public OpenAPI endpoint: https://normalize.cancervariants.org/variation

Installing with pip:

pip install variation-normalizer

The variation-normalization repo depends on VRS models, and therefore each variation-normalizer package on PyPI uses a particular version of VRS. The correspondences between packages may be summarized as:

variation-normalization branch variation-normalizer version gene-normalizer version VRS version
main 0.6.X 0.1.X 1.X.X
staging >=0.8.X >=0.3.X 2.0-alpha

About

Variation Normalization works by using four main steps: tokenization, classification, validation, and translation. During tokenization, we split strings on whitespace and parse to determine the type of token. During classification, we specify the order of tokens a classification can have. We then do validation checks such as ensuring references for a nucleotide or amino acid matches the expected value and validating a position exists on the given transcript. During translation, we return a VRS Allele object.

Variation Normalization is limited to the following types of variants:

  • HGVS expressions and text representations (ex: BRAF V600E):
    • protein (p.): substitution, deletion, insertion, deletion-insertion
    • coding DNA (c.): substitution, deletion, insertion, deletion-insertion
    • genomic (g.): substitution, deletion, ambiguous deletion, insertion, deletion-insertion, duplication
  • gnomAD-style VCF (chr-pos-ref-alt, ex: 7-140753336-A-T)
    • genomic (g.): substitution, deletion, insertion

Variation Normalizer accepts input from GRCh37 or GRCh8 assemblies.

We are working towards adding more types of variations, coordinates, and representations.

Endpoints

/to_vrs

Returns a list of validated VRS Variations.

/normalize

Returns a VRS Variation aligned to the prioritized transcript. The Variation Normalizer relies on Common Operations On Lots-of Sequences Tool (cool-seq-tool) for retrieving the prioritized transcript data. More information on the transcript selection algorithm can be found here.

If a genomic variation query is given a gene (E.g. BRAF g.140753336A>T), the associated cDNA representation will be returned. This is because the gene provides additional strand context. If a genomic variation query is not given a gene, the GRCh38 representation will be returned.

Developer Instructions

Clone the repo:

git clone https://github.com/cancervariants/variation-normalization.git
cd variation-normalization

For a development install, we recommend using Pipenv. See the pipenv docs for direction on installing pipenv in your compute environment.

Once installed, from the project root dir, just run:

pipenv shell
pipenv update && pipenv install --dev

Backend Services

Variation Normalization relies on some local data caches which you will need to set up. It uses pipenv to manage its environment, which you will also need to install.

Gene Normalizer

Variation Normalization relies on data from Gene Normalization. You must load all sources and merged concepts.

You must also have Gene Normalization's DynamoDB running in a separate terminal for the application to work.

For more information about the gene-normalizer and how to load the database, visit the README.

SeqRepo

Variation Normalization relies on seqrepo, which you must download yourself.

Variation Normalizer uses seqrepo to retrieve sequences at given positions on a transcript.

From the root directory:

pip install seqrepo
sudo mkdir /usr/local/share/seqrepo
sudo chown $USER /usr/local/share/seqrepo
seqrepo pull -i 2021-01-29  # Replace with latest version using `seqrepo list-remote-instances` if outdated

If you get an error similar to the one below:

PermissionError: [Error 13] Permission denied: '/usr/local/share/seqrepo/2021-01-29._fkuefgd' -> '/usr/local/share/seqrepo/2021-01-29'

You will want to do the following:
(Might not be ._fkuefgd, so replace with your error message path)

sudo mv /usr/local/share/seqrepo/2021-01-29._fkuefgd /usr/local/share/seqrepo/2021-01-29
exit

Use the SEQREPO_ROOT_DIR environment variable to set the path of an already existing SeqRepo directory. The default is /usr/local/share/seqrepo/latest.

UTA

Variation Normalizer also uses Common Operations On Lots-of Sequences Tool (cool-seq-tool) which uses UTA as the underlying PostgreSQL database.

The following commands will likely need modification appropriate for the installation environment.

  1. Install PostgreSQL

  2. Create user and database.

    createuser -U postgres uta_admin
    createuser -U postgres anonymous
    createdb -U postgres -O uta_admin uta
  3. To install locally, from the variation/data directory:

export UTA_VERSION=uta_20210129.pgd.gz
curl -O http://dl.biocommons.org/uta/$UTA_VERSION
gzip -cdq ${UTA_VERSION} | grep -v "^REFRESH MATERIALIZED VIEW" | psql -h localhost -U uta_admin --echo-errors --single-transaction -v ON_ERROR_STOP=1 -d uta -p 5433
UTA Installation Issues

If you have trouble installing UTA, you can visit these two READMEs.

Connecting to the UTA database

To connect to the UTA database, you can use the default url (postgresql://uta_admin@localhost:5433/uta/uta_20210129). If you do not wish to use the default, you must set the environment variable UTA_DB_URL which has the format of driver://user:pass@host:port/database/schema.

Starting the Variation Normalization Service Locally

gene-normalizers dynamodb and the uta database must be running.

To start the service, run the following:

uvicorn variation.main:app --reload

Next, view the OpenAPI docs on your local machine: http://127.0.0.1:8000/variation

Init coding style tests

Code style is managed by Ruff and checked prior to commit.

Check style with ruff:

python3 -m ruff format . && python3 -m ruff check --fix .

We use pre-commit to run conformance tests.

This ensures:

  • Style correctness
  • No large files
  • AWS credentials are present
  • Private key is present

Pre-commit must be installed before your first commit. Use the following command:

pre-commit install

Testing

From the root directory of the repository:

pytest tests/