8  References

Tips and links to potentially useful learning resources.

8.1 Tool Tips - Git

• Version control with Git: Software Carpentries course.

• How to undo (almost) anything: Tips on how to undo various actions in git.

• Accidentally delete a tracked file?

  git restore <filename>

• Do not add large ( > 100MB ) files.

• Never commit sensitive information to a git repository e.g.,

  • usernames and passwords
  • API keys

8.2 Tool Tips - Docker

• Reproducible Computational Environments Using Containers: Introduction to Docker: Software Carpentries course.

• Dockerfile best practices

• Use existing containers where possible:

  • Biocontainers Registry: Containerized conda packages.
  • Rocker: R in Docker.
    • Rocker TidyModels: R + Rstudio + tidyverse + tidymodels + common ML packages.
  • Python Data science: Common python dependencies for data science workflows.

• Be careful not to commit sensitive data into a container.

• Docker compose: Add parameters and volumes to a docker-compose.yml file to reproduce a Docker ecosystem (start multiple containers with the same settings each time).

8.3 Tool Tips - Apptainer

• Reproducible computational environments using containers: Introduction to Singularity: Software Carpentries course (Work in progress).

• Build custom images in Docker first, for increased portabilty.

• Use existing containers where possible:

  • Biocontainers Registry: Containerized conda packages.
  • Rocker: R in Docker.
    • Rocker TidyModels: R + Rstudio + tidyverse + tidymodels + common ML packages.
  • Python Data science: Common python dependencies for data science workflows.

• Be careful not to commit sensitive data into a container.

8.4 Tool Tips - Nextflow

• Nextflow Training from Seqera: Training from writers of Nextflow.

• Introduction to Bioinformatics workflows with Nextflow and nf-core: Software Carpentries Course (Work in progress).

• RNA-seq Workflow Tutorial: Tutorial writing an RNA-seq workflow to introduce Nextflow fundamentals.

• GATK Workflow Tutorial: Tutorial writing a GATK workflow to introduce Nextflow fundamentals.

• Use toy examples to test Nextflow contructs.

  Channel.of( 
      [ "hi",  "there", 1 ], 
      [ "hi",  "there", 1 ],
      [ "see", "me",    2 ], 
      [ "see", "how",   2 ]
    )
    .unique()
    .groupTuple( by:[0,2] )
    .view()

• Nextflow documentation: Nextflow Workflow writers best friend. A description of all the available Nextflow functionality.

8.5 Tool Tips - Conda

• Introduction to Conda for (Data) Scientists: Software carpentries course.
• Do not install anything into the base environment.
• I recommend mamba (install it into a conda env).