Reproducible Data Workflows with Snakemake [ResBaz]

Does your data analysis require several steps across various software? Do you need to run the same analysis repeatedly and reproducibly? These common scenarios in digital research can lead to complex manual processes with tedious file handling and a high chance of human error. Workflow languages solve these issues by automating your data analysis with code. They provide reproducibility by ensuring each workflow runs consistently every time. They allow you to organise your software, inputs, outputs and logging for clear versioning, reporting, and results. They are even self documenting, providing a clear illustration of how your whole workflow fits together. Finally they allow you to scale your workflows up for running on HPC such as REANNZ HPC. A well-defined workflow means you can set your full data analysis running and go make a cup of tea knowing you’ll come back to accurate outputs and clear logs. In this workshop, we will work through an introduction to Snakemake, a workflow language with its basis in the popular programming language, Python. This Workshop is intended for anyone who has several steps in their data analysis workflow, particularly when many different software tools are involved. Basic command line experience as provided in "Introduction To the Command Line" is highly recommended, but no other programming experience is required.