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
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.
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