Lessons¶

Nine core lessons, taken in order, build from open-science principles to reproducible, scalable computing; an extra lesson covers workflow managers.

Open science and data¶

• Introduction to Open Science - What open science is, its six pillars (open access, data, educational resources, methodology, peer review, and source software), and why researchers practice it.
• Data Management and Documentation - The data life cycle, FAIR and CARE principles, data management plans, licensing, and tools for documenting research projects.

Talking to machines¶

• How to Talk to Computers - An introduction to the Unix shell: navigating directories, working with files, and automating tasks with shell scripts.
• How to Talk to LLMs - How large language models work, prompt-engineering techniques such as priming, chaining, and the CRAFT framework, and AI coding tools for research.

Reproducibility¶

• Reproducibility I: Version Control - Version control with Git and GitHub: repositories, the commit-push-pull cycle, branching, pull requests, and handling large files.
• Reproducibility II: Software Environments - Computing environments, the system PATH, and using Conda, pip, and renv to create, share, and reproduce software environments.
• Reproducibility III: Containers - Why containers matter for reproducible research, core Docker commands, running an example container, and writing, building, and pushing a Dockerfile.

Remote computing¶

• Remote Computing: HPC - Logging in to a high-performance computing cluster, submitting Slurm batch and interactive jobs, loading modules, and running Apptainer containers.
• Remote Computing: CyVerse - An overview of the CyVerse cloud platform: the Discovery Environment, Data Store, Data Commons, analysis apps, and the analysis dashboard.

Beyond the core lessons¶

• AI Models & Where to Find Them - Lesson 10 is taught from the separate GPT 101 generative-AI workshop: finding, running, and fine-tuning AI models for research.
• Extra: Workflows with Nextflow - A comparison of the Nextflow and Snakemake workflow managers and a hands-on RNA-seq tutorial that runs a Nextflow pipeline in a Conda environment.