On occasion it’s necessary to build very large container. The build time limit (BTL) with Sylabs is 1 hr, so if your container requires packages with lots of dependecies you’ll often exceed the BTL. One way around this is to build the container in smaller chunks, and layering up the packages bit by bit.

Below is a definition file I’ve been using to build a container for Seurat 5. This version of Seurat has new functionality to more efficently analyse large datasets (1M cells) and requires several susidiary packages to facilitate this.

Definition file

``` Bootstrap: docker From: bioconductor/bioconductor_docker:devel %labels Version v0.0.1 %help This is a container to run Seurat 5 on Hawk %post # Update the image apt update apt upgrade -y # for igraph apt install -y glpk-utils libglpk-dev # for sctransform apt install -y libicu-dev # for BPCells apt install -y libhdf5-dev # Install R packages R --no-echo -e 'remotes::install_github("bnprks/BPCells")' R --no-echo -e 'BiocManager::install("glmGamPoi")' R --no-echo -e 'install.packages("RPresto")' R --no-echo -e 'install.packages("Seurat")' R --no-echo -e 'setRepositories(ind=1:3)' # Needed to automatically install Bioconductor dependencies for Signac R --no-echo -e 'install.packages(c("R.utils", "Signac"))' R --no-echo -e 'remotes::install_github("satijalab/seurat-wrappers")' R --no-echo -e 'remotes::install_github("satijalab/azimuth")' R --no-echo -e 'BiocManager::install(c("scuttle", "scater"))' R --no-echo -e 'install.packages("scCustomize")' R --no-echo -e 'install.packages("readxl")' apt clean ```

Instead of running the entire thing at once, which exceeds the BTL, I built the container up until the R --no-echo -e 'install.packages("Seurat")' line first:

Layer 1

``` Bootstrap: docker From: bioconductor/bioconductor_docker:devel %labels Version v0.0.1 %help This is a container to run Seurat 5 on Hawk %post # Update the image apt update apt upgrade -y # for igraph apt install -y glpk-utils libglpk-dev # for sctransform apt install -y libicu-dev # for BPCells apt install -y libhdf5-dev # Install R packages R --no-echo -e 'remotes::install_github("bnprks/BPCells")' R --no-echo -e 'BiocManager::install("glmGamPoi")' R --no-echo -e 'install.packages("RPresto")' R --no-echo -e 'install.packages("Seurat")' apt clean ```

This took about 45 minutes to build. I then used that container as the root conatiner and added the remianing packages to it.

Layer 2

``` Bootstrap: library From: library://dazcam/containers/seurat5a:latest %labels Version v0.0.1 %help This is a container to run Seurat 5 on Hawk %post # Install R packages R --no-echo -e 'setRepositories(ind=1:3)' # Needed to automatically install Bioconductor dependencies for Signac R --no-echo -e 'install.packages(c("R.utils", "Signac"))' R --no-echo -e 'remotes::install_github("satijalab/seurat-wrappers")' R --no-echo -e 'remotes::install_github("satijalab/azimuth")' R --no-echo -e 'BiocManager::install(c("scuttle", "scater"))' R --no-echo -e 'install.packages("scCustomize")' R --no-echo -e 'install.packages("readxl")' ```

A few other points to mention here.

First, notice the apt commands in layer A. Apt is a package manager for linux distributions. From time to time, R packages require additional software to be added to the underlying linux kernal. These can be added as needed using apt update / upgrade / install.

Second, notice the difference in the headers for layer A and layer B. Layer A uses the bioconductor base container as the input container from Docker Hub, whereas layer B uses the Layer A container stored in my personal singularity library.

---

Move back to Build a basic container, or back to index page.