gentroutils

Set of Command Line Interface tools to process Open Targets Genetics GWAS data.

Installation

pip install gentroutils

Available commands

To see all available commands after installation run

gentroutils --help

Usage

To run a single step run

uv run gentroutils -s gwas_catalog_release  # After cloning the repository
gentroutils -s gwas_catalog_release -c otter_config.yaml # When installed by pip

The gentroutils repository uses the otter framework to build the set of tasks to run. The current implementation of tasks can be found in the config.yaml file in the root of the repository. To run gentroutils installed via pip you need to define the otter config that looks like the config.yaml file.

---
work_path: ./work
log_level: DEBUG
scratchpad:
steps:
  gwas_catalog_release:
    - name: crawl release metadata
      stats_uri: "https://www.ebi.ac.uk/gwas/api/search/stats"
      destination_template: "gs://gwas_catalog_inputs/gentroutils/{release_date}/stats.json"
      promote: "true"
    - name: fetch associations
      stats_uri: "https://www.ebi.ac.uk/gwas/api/search/stats"
      source_template: "ftp://ftp.ebi.ac.uk/pub/databases/gwas/releases/{release_date}/gwas-catalog-associations_ontology-annotated.tsv"
      destination_template: "gs://gwas_catalog_inputs/gentroutils/{release_date}/gwas_catalog_associations_ontology_annotated.tsv"
      promote: true
    - name: fetch studies
      stats_uri: "https://www.ebi.ac.uk/gwas/api/search/stats"
      source_template: "ftp://ftp.ebi.ac.uk/pub/databases/gwas/releases/{release_date}/gwas-catalog-download-studies-v1.0.3.1.txt"
      destination_template: "gs://gwas_catalog_inputs/gentroutils/{release_date}/gwas_catalog_download_studies.tsv"
      promote: true
    - name: fetch ancestries
      stats_uri: "https://www.ebi.ac.uk/gwas/api/search/stats"
      source_template: "ftp://ftp.ebi.ac.uk/pub/databases/gwas/releases/{release_date}/gwas-catalog-download-ancestries-v1.0.3.1.txt"
      destination_template: "gs://gwas_catalog_inputs/gentroutils/{release_date}/gwas_catalog_download_ancestries.tsv"
      promote: true
    - name: curation study
      requires:
        - fetch studies
      previous_curation: gs://gwas_catalog_inputs/curation/latest/curated/GWAS_Catalog_study_curation.tsv
      studies: gs://gwas_catalog_inputs/gentroutils/latest/gwas_catalog_download_studies.tsv
      destination_template: gs://gwas_catalog_inputs/gentroutils/curation/{release_date}/GWAS_Catalog_study_curation.tsv
      summary_statistics_glob: gs://gwas_catalog_inputs/raw_summary_statistics/*.h.tsv.gz
      promote: true

Available tasks

The list of tasks (defined in the config.yaml file) that can be run are:

Crawl release metadata

- name: crawl release metadata
      stats_uri: "https://www.ebi.ac.uk/gwas/api/search/stats"
      destination_template: "gs://gwas_catalog_inputs/gentroutils/{release_date}/stats.json"
      promote: "true"

This task fetches the latest GWAS Catalog release metadata from the https://www.ebi.ac.uk/gwas/api/search/stats endpoint and saves it to the specified destination.

Fetch associations

- name: fetch associations
      stats_uri: "https://www.ebi.ac.uk/gwas/api/search/stats"
      source_template: "ftp://ftp.ebi.ac.uk/pub/databases/gwas/releases/{release_date}/gwas-catalog-associations_ontology-annotated.tsv"
      destination_template: "gs://gwas_catalog_inputs/gentroutils/{release_date}/gwas_catalog_associations_ontology_annotated.tsv"
      promote: true

This task fetches the GWAS Catalog associations file from the specified FTP server and saves it to the specified destination.

Fetch studies

- name: fetch studies
      stats_uri: "https://www.ebi.ac.uk/gwas/api/search/stats"
      source_template: "ftp://ftp.ebi.ac.uk/pub/databases/gwas/releases/{release_date}/gwas-catalog-download-studies-v1.0.3.1.txt"
      destination_template: "gs://gwas_catalog_inputs/gentroutils/{release_date}/gwas_catalog_download_studies.tsv"
      promote: true

This task fetches the GWAS Catalog studies file from the specified FTP server and saves it to the specified destination.

Fetch ancestries

- name: fetch ancestries
      stats_uri: "https://www.ebi.ac.uk/gwas/api/search/stats"
      source_template: "ftp://ftp.ebi.ac.uk/pub/databases/gwas/releases/{release_date}/gwas-catalog-download-ancestries-v1.0.3.1.txt"
      destination_template: "gs://gwas_catalog_inputs/gentroutils/{release_date}/gwas_catalog_download_ancestries.tsv"
      promote: true

This task fetches the GWAS Catalog ancestries file from the specified FTP server and saves it to the specified destination.

Curation

- name: curation study
      requires:
        - fetch studies
      previous_curation: gs://gwas_catalog_inputs/curation/latest/curated/GWAS_Catalog_study_curation.tsv
      studies: gs://gwas_catalog_inputs/gentroutils/latest/gwas_catalog_download_studies.tsv
      destination_template: gs://gwas_catalog_inputs/curation/{release_date}/raw/gwas_catalog_study_curation.tsv
      summary_statistics_glob: gs://gwas_catalog_inputs/raw_summary_statistics/*.h.tsv.gz
      promote: true

This task is used to build the GWAS Catalog curation file that is later used as a template for manual curation. It requires the fetch studies task to be completed before it can run. This is due to the fact that the curation file is build based on the list of studies fetched from download studies file.

Curation process

The base of the curation process for GWAS Catalog data is defined in the docs/gwas_catalog_curation.md. The original solution uses R script to prepare the data for curation and then manually curates the data. The solution proposed in the curation task automates the preparation of the data for curation and provides a template for manual curation. The manual curation process is still required, but the data preparation is automated.

The automated process includes:

1. Reading download studies file with the list of studies that are currently comming from the latest GWAS Catalog release.
2. Reading previous curation file that contains the list of the curated studies from the previous release.
3. Listing all synced summary statistics files from the summary_statistics_glob parameter to identify which studies have summary statistics available. Note that this can be more then the list of studies in the download studies file as syncing also involves the unpublished studies.
4. Comparing the three datasets with following logic:
   • In case the study is present in the previous curation and download studies, the study is marked as curated
   • In case the study is present in the download studies but not in the previous curation, the study is marked as to_curate or has_no_sumstats depending on the presence of summary statistics files
   • In case the study is present in the previous curation but not in the download studies, the study is marked as removed
5. The output of the curation process is a file that contains the list of studies with their status (curated, new, removed) and the fields that are required for manual curation. The output file is saved to the destination_template path specified in the task configuration. The file is saved under gs://gwas_catalog_inputs/curation/{release_date}/raw/gwas_catalog_study_curation.tsv path.
6. The output file is then promoted to the latest release path gs://gwas_catalog_inputs/curation/latest/raw/gwas_catalog_study_curation.tsv so that it can be used for manual curation.
7. The manual curation process is then performed on the gs://gwas_catalog_inputs/curation/latest/raw/gwas_catalog_study_curation.tsv file. The manual curation process is not automated and requires manual intervention. The output from the manual curation process should be saved then to the gs://gwas_catalog_inputs/curation/latest/curated/GWAS_Catalog_study_curation.tsv and gs://gwas_catalog_inputs/curation/{release_date}/curated/GWAS_Catalog_study_curation.tsv file. This file is then used for the Open Targets Staging Dags.

Contribute

To be able to contribute to the project you need to set it up. This project runs on:

• python 3.13
• uv (dependency manager)

To set up the project run

make dev

The command will install above dependencies (initial requirements are curl and bash) if not present and install all python dependencies listed in pyproject.toml. Finally the command will install pre-commit hooks required to be run before the commit is created.

The project has additional dev dependencies that include the list of packages used for testing purposes. All of the dev dependencies are automatically installed by uv.

To see all available dev commands

Run following command to see all available dev commands

make help

Manual testing of CLI module

To check CLI execution manually you need to run

uv run gentroutils

This software was developed as part of the Open Targets project. For more information please see: http://www.opentargets.org