Docker images for reproducible single-cell data analyses

Maintainer: António Sousa ([email protected])

Contents

• Build image

• Run container

Build image

1. Main: Build & push the image to personal docker hub (aggode) - compatible only with amd64:

docker login -u aggode # type password

docker build -f Dockerfile-amd64 -t sctoolkit .

docker image tag sctoolkit aggode/sctoolkit:latest

docker image push aggode/sctoolkit:latest

2. Alternative: Build & push the image to personal docker hub (aggode) - compatible amd64 & arm64:

ongoing: can't install system library llvm-10 w/ base image rocker/rstudio:latest-daily):

docker run --privileged --rm tonistiigi/binfmt --install all

docker buildx create --name mybuilder

docker buildx use mybuilder

docker buildx inspect --bootstrap

docker buildx build --no-cache --push --platform linux/amd64,linux/arm64 -t aggode/sctoolkit:latest -f Dockerfile-cross-platform .

Run container

Singularity

singularity pull docker://aggode/sctoolkit:latest