gwas-tools

Pipelines for common use-cases when dealing with GWAS datasets, from preprocessing to biomarker discovery. All dependencies are provided in a homonymous Docker image.

Installation

Simply clone the repository and add the bin/ folder to your path:

git clone git@github.com:hclimente/gwas-tools.git
export PATH=$PATH:$PWD/gwas-tools/bin

Install also the two dependencies:

• Nextflow
• Choose one of:
  • Docker
  • Singularity

Each pipeline uses multiple tools and libraries, which I compile in a Docker image. Those that can be freely distributed are available in hclimente/gwas-tools. You will need to include the remaining ones into a new Docker image (hclimente/gwas-tools-extra) by running:

make docker

Then, simply add '-with-docker hclimente/gwas-tools[-extra]' or '-with-singularity hclimente/gwas-tools[-extra]' when launching a pipeline.

Functions

• GWAS
  • Data preprocessing
  • Network-guided GWAS
• Epistasis detection
  • Network-guided epistasis detection

GWAS

Data preprocessing

• Impute a dataset: impute --bfile test/data/gwas --strand_info test/data/strand_info.txt --population EUR -with-docker hclimente/gwas-tools
• Run VEGAS2: vegas2.nf --snp_association test/data/assoc.chisq --bfile_ld_controls test/data/gwas --vegas2_params '-top 10' -with-docker hclimente/gwas-tools
• Map SNPs to GENCODE genes: snp2gene.nf --bim test/data/gwas.bim -with-docker hclimente/gwas-tools

Network-guided GWAS

We adapted and benchmarked multiple algorithms for the detection of SNPs associated to a phenotype. If you use any of the following algorithms, please cite the original algorithm's article, and the following article:

Climente-González H, Lonjou C, Lesueur F, GENESIS study group, Stoppa-Lyonnet D, et al. (2021) Boosting GWAS using biological networks: A study on susceptibility to familial breast cancer. PLOS Computational Biology 17(3): e1008819. https://doi.org/10.1371/journal.pcbi.1008819

The available methods are:

• dmGWAS: dmgwas.nf --scores test/data/vegas2.tsv --edgelist test/data/edgelist.tsv -with-docker hclimente/gwas-tools
• heinz: heinz.nf --scores test/data/vegas2.tsv --edgelist test/data/edgelist.tsv -with-docker hclimente/gwas-tools
• HotNet2: hotnet2.nf --scores test/data/vegas2.tsv --edgelist test/data/edgelist.tsv --network_permutations 2 --heat_permutations 2 -with-docker hclimente/gwas-tools-extra
• LEAN: lean.nf --scores test/data/vegas2.tsv --edgelist test/data/edgelist.tsv -with-docker hclimente/gwas-tools
• SConES: scones.nf --bfile test/data/gwas --network gi --snp2gene test/data/snp2gene.tsv --edgelist test/data/edgelist.tsv -with-docker hclimente/gwas-tools
• Sigmod: sigmod.nf --scores test/data/vegas2.tsv --edgelist test/data/edgelist.tsv --nmax 6 --maxjump 1 -with-docker hclimente/gwas-tools

Network-guided GWAS using stability selection

If you use this algorithm, please cite the following article:

Climente-González H et al. (2023) Network-guided GWAS using stability selection. Accepted in STAR Protocols.

Usage:

stable_network_gwas.nf \
    --bfile test/data/gwas \
    --edgelist test/data/edgelist.tsv \
    --sigmod_nmax 6 \
    --sigmod_maxjump 1 \
    -with-docker hclimente/gwas-tools

Epistasis detection

Network-guided epistasis detection

We developed a modular method to discover epistasis along the edges of a biological network. This facilitates the interpretability of the results and allows to find interactions that would otherwise be overcome by the multiple test burden. If you use this algorithm, please cite the following article:

Duroux D, Climente-González H, Azencott C-A, Van Steen K (2021) Interpretable network-guided epistasis detection. In press. https://doi.org/10.1101/2020.09.24.310136

Usage:

network_epistasis.nf \
    --bfile test/data/gwas \
    --tab2  test/data/interactions.tab2 \
    --snp2gene test/data/snp2gene.tsv \
    --nperm 10