drake-bioinformatics

This repo contains slides for a talk about the {drake} package at a London Bioinformatics Meetup on 30 January 2020.

The materials are:

• the slides in your browser
• the source for the slides (this repo)
• an .R script file of all the {drake} code

Workflow management

The {drake} package by Will Landau is an R-focused solution for managing your analytical workflows.

What does that mean? {drake} ‘remembers’ the relationships between objects in your workflow. When you update your analysis, {drake} makes sure that only the impacted objects are re-run. This means you don’t have to recreate everything from scratch each time.

Resources

Official {drake} materials:

• rOpenSci site
• the user manual
• an rOpenSci community call
• learndrake in the cloud
• drakeplanner Shiny app
• launch drake examples in the cloud
• source on GitHub
• the CRAN listing

Great talks by other people include:

• the ‘reproducible workflows with the drake R package’ talk by Kirill Müller (14 April 2019)
• the ‘reproducible data workflows’ talk by Garrick Aden-Buie (19 July 2019)
• a use case for tracking NYC fires with Twitter and Google by Amanda Dobbyn (19 June 2019)

I’ve written about {drake} before:

• in a blog post called can {drake} RAP?
• in a Coffee & Coding presentation with an accompanying {drake} demo repo
• in another Coffee & Coding presentation about three things to improve reproducibility
• as one of my packages that sparked joy in 2019