dcd-map: Map MaveDB data to computable and interoperable variant objects

This library implements a novel method for mapping MaveDB scoreset data to GA4GH Variation Representation Specification (VRS 2.0) objects, enhancing interoperability for genomic medicine applications. See Arbesfeld et. al. (2023) for a preprint edition of the mapping manuscript, or download the resulting mappings directly.

Prerequisites

• Universal Transcript Archive (UTA): see README for setup instructions. Users with access to Docker on their local devices can use the available Docker image; otherwise, start a relatively recent (version 14+) PostgreSQL instance and add data from the available database dump.
• SeqRepo: see README for setup instructions. The SeqRepo data directory must be writeable; see specific instructions here for more.
• Gene Normalizer: see documentation for data setup instructions.
• blat: Must be available on the local PATH and executable by the user. Otherwise, its location can be set manually with the BLAT_BIN_PATH env var. See the UCSC Genome Browser FAQ for download instructions.

Installation

Install from PyPI:

python3 -m pip install dcd-mapping

Usage

Use the dcd-map command with a scoreset URN, eg

$ dcd-map urn:mavedb:00000083-c-1

Output is saved in the format <URN>_mapping_results_<ISO datetime>.json in the directory specified by the environment variable MAVEDB_STORAGE_DIR, or ~/.local/share/dcd-mapping by default.

Use dcd-map --help to see other available options.

Notebooks

Notebooks for manuscript data analysis and figure generation are provided within notebooks/analysis. See notebooks/analysis/README.md for more information.

Development

Clone the repo

git clone https://github.com/ave-dcd/dcd_mapping
cd dcd_mapping

Create and activate a virtual environment

python3 -m virtualenv venv
source venv/bin/activate

Install as editable and with developer dependencies

python3 -m pip install -e '.[dev,tests]'

Add pre-commit hooks

pre-commit install

Run tests with pytest

pytest