Variant Classification DB

Introduction

A database to hold information about variant classifications done within the laboratory.

The main purpose of this database is to store the results of classifying variants using the ACMG guidelines [1] and CNVs using the ACMG CNV guidelines [2].

Install and Setup

Software Setup

Works on Centos 6/7

The software is a Django application using Python 3. It is reccomended that the software be deployed in a conda virtual environment.

First install Conda/Miniconda from [3]. Then type the following commands in your terminal to install and setup the application.

Optionally collect the gene annotations from the variant_filtering_app/data_sets/gene_data.csv

git clone https://github.com/AWGL/variant_filtering_app.git

Download main repository

git clone https://github.com/josephhalstead/variant_classification_DB.git

cd variant_classification_DB

conda env create -f env/acmg_db.yaml

source activate acmg_db

python manage.py migrate

python manage.py makemigrations acmg_db

python manage.py migrate

python manage.py createsuperuser

python manage.py loaddata acmg_questions.json

python manage.py loaddata CNV_Gain_ACMG_questions.json

python manage.py loaddata CNV_Loss_ACMG_questions.json

Optionally add gene data:

python manage.py load_genes --gene_file variant_filtering_app/data_sets/gene_data.csv

Static files which are viewed on the webpage also need to be moved to the static directory:

mv files/* static/files/

Other Resources

• Reference Genome - either Human Reference GRCh37 or Human Reference GRCh38
• VEP Cache for GRCh37 - http://ftp.ensembl.org/pub/release-100/variation/indexed_vep_cache/homo_sapiens_refseq_vep_100_GRCh37.tar.gz
• VEP Cache for GRCh38 - http://ftp.ensembl.org/pub/release-104/variation/indexed_vep_cache/homo_sapiens_refseq_vep_104_GRCh38.tar.gz

Settings

There are also some options in mysite/settings.py that may need to be changed:

SECRET_KEY = 'xxxxxx'

ALLOWED_HOSTS = ['127.0.0.1']

MUTALYZER_URL = 'https://mutalyzer.nl/services/?wsdl'

MUTALYZER_BUILD = 'hg19'

REFERENCE_GENOME = '/media/sf_Documents/genomics_resources/refs/human_g1k_v37.fasta'

VEP_CACHE = '/media/sf_Documents/genomics_resources/vep/'

VEP_TEMP_DIR = 'temp/'

Set these to your preferred settings or use the default. The secret key should be changed if running in production.

Test

There are some tests for the acmg classifier as well as checking all the views work.

python manage.py test --keepdb

Note - encoding error when running tests: If run without the keepdb flag the encoding will default to LATIN9, which won't be able to load the example data fixtures, and then the database will be deleted. If the test database is kept with the keepdb flag, you can log into the psql terminal and change the encoding to UTF8 with the command update pg_database set encoding = pg_char_to_encoding('UTF8') where datname = 'test_variant_classification_db', and then re-run the tests again.

Run

To run using the development server:

python manage.py runserver

Saving JSON file of ACMG criteria

If the ACMG criteria are changed, a JSON file should be saved into the fixtures folder so that a new database can be initialised with the same criteria.

To do this, run the command python manage.py dumpdata acmg_db.classificationquestion > <date>_acmg_questions.json

References

[1] Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology https://www.acmg.net/docs/standards\_guidelines\_for\_the\_interpretation\_of\_sequence\_variants.pdf

[2] Technical standards for the interpretation and reporting of constitutional copy number variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics (ACMG) and the Clinical Genome Resource (ClinGen) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7313390/

[3] https://conda.io/miniconda.html