zktuong/sc-dandelion repository overview

⁠sc-dandelion

Container for dandelion⁠, a Python package for analysing single-cell BCR/TCR (V(D)J) data from 10x Genomics 5' VDJ, BD Rhapsody, Parse Biosciences, SeekGene and other AIRR-compatible sources.

The image comes with everything needed for V(D)J preprocessing already installed: dandelion, IgBLAST, BLAST, Change-O, TIgGER and pre-built germline reference databases (IMGT and others).

⁠Which tag?

Every release is published in two forms:

⁠Pull

Apptainer: ready-made .sif (fastest, recommended on HPC)

apptainer pull sc-dandelion_latest.sif oras://docker.io/zktuong/sc-dandelion:latest-sif

Apptainer: build the .sif from the Docker image (takes a few minutes and a few GB of temporary space)

apptainer pull docker://zktuong/sc-dandelion:latest

Both give you sc-dandelion_latest.sif. To pin a release, use e.g. v0.5.0-sif or v0.5.0 instead of latest-sif/latest (see the Tags tab). Using SingularityCE? Replace apptainer with singularity; the commands are otherwise identical.

Docker / Podman

docker pull zktuong/sc-dandelion:latest

If you hit Docker Hub's pull rate limit (common on shared HPC login nodes), the same images are mirrored on GitHub. Replace docker.io/zktuong with ghcr.io/tuonglab:

apptainer pull sc-dandelion_latest.sif oras://ghcr.io/tuonglab/sc-dandelion:latest-sif

⁠Usage

Run V(D)J preprocessing from inside your data folder:

apptainer run -B $PWD sc-dandelion_latest.sif dandelion-preprocess

For multiple samples, provide a metadata .csv:

apptainer run -B $PWD sc-dandelion_latest.sif dandelion-preprocess --meta meta.csv

See all options:

apptainer run -B $PWD sc-dandelion_latest.sif dandelion-preprocess --help

Change-O clonotype calling:

apptainer run -B $PWD sc-dandelion_latest.sif changeo-clonotypes --help

Interactive shell with all tools available:

apptainer shell --writable-tmpfs -B $PWD sc-dandelion_latest.sif

With Docker:

docker run --rm -v "$PWD":/data -w /data --user "$(id -u):$(id -g)" \
    zktuong/sc-dandelion:latest dandelion-preprocess --meta meta.csv

⁠Moved from Sylabs Cloud

This image used to be hosted at library://kt16/default/sc-dandelion. The Sylabs Cloud Library has been decommissioned, so please update any scripts to use the addresses above.

⁠Links

• Documentation: https://sc-dandelion.readthedocs.io/⁠
• Container preprocessing tutorial: https://sc-dandelion.readthedocs.io/en/latest/notebooks/Q1-container-preprocessing.html⁠
• Try it in Google Colab: https://colab.research.google.com/github/tuonglab/dandelion/blob/master/container/dandelion_colab.ipynb⁠
• Source code and issues: https://github.com/tuonglab/dandelion⁠
• Python package: https://pypi.org/project/sc-dandelion/⁠

⁠Citation

If you use dandelion, please cite:

Suo, C., Polanski, K., Dann, E. et al. Dandelion uses the single-cell adaptive immune receptor repertoire to explore lymphocyte developmental origins. Nat Biotechnol (2023). https://doi.org/10.1038/s41587-023-01734-7⁠