/growthis

Primary LanguageROtherNOASSERTION

The growthis package

Introduction

This package can be used to view, edit and analyse Varioskan growth experiment data generated in a 96-well Excel format.

System Requirements

You need to have these two or three tools installed.

  1. R (version >4.1). Download from https://cran.r-project.org/

  2. RStudio (version >2021.09.01). Download from https://www.rstudio.com/products/rstudio/download/ (choose RStudio Desktop Free).

  3. [Windows only] On Windows you will need RTools, which can be downloaded at https://cran.r-project.org/bin/windows/Rtools/. Choose the version that is advised for your R version. You can check the R version in RStudio: at startup it will display a message in the Console pane:

    R version 4.1.2 (2021-11-01) -- "Bird Hippie"
Copyright (C) 2021 The R Foundation for Statistical Computing
Platform: aarch64-apple-darwin21.1.0 (64-bit)
...

    So in this example the R version is 4.1.2 and I should download RTools version 4.0 (if I were on a Windows machine).

If you have an R version below 4.0 or RStudio below 2021.0 I suggest you upgrade first (same as the steps above). Minor RStudio updates can be done from within RStudio itself: via Help → Check for updates. Have a look at https://www.linkedin.com/pulse/3-methods-update-r-rstudio-windows-mac-woratana-ngarmtrakulchol for more info.

Getting started

First-time installation of the growthis package

Once you have R and RStudio in place you need to set up the R environment to install and work with the growthis package and app. Usually in R you will use the function install.packages() to add external libraries to R (a library is a set of functions and sometimes also data). This function will look in the central repository for all checked and registered packages: CRAN. Not all libraries are hosted in CRAN however, and growthis is one of them, and to get these installed we need other means. One of these is via devtools::install_github(). devtools is a package that is hosted on CRAN.

Here are the steps (with some explanation):

  1. Start RStudio and go to the Console panel. All commands below need to be entered in the Console.

  2. Type install.packages("devtools") and press Enter. This will install the package devtools in the running R distribution if not already present.

  3. Type library(devtools) and press Enter. This will load devtools in the current R session.

  4. Type install_github("MichielNoback/growthis") and press Enter. This will install package growthis.
    R will probably "ask" you to update some (or quite a few) packages. You should preferably select All (option 1). This may take a while on some systems. Don't be scared of all the text running by. However, if the stream stops and it say "Error..." something has gone wrong and you should ask a more experienced colleague for help.

  5. Test it. Type library(growthis) and press Enter. This will load the growthis package. If it prints Error in library(growthis) : there is no package called ‘growthis’ something has gone wrong in step 4 (or before). Try the installation again, reading carefully any warning and error messages. If that fails, try to get help at this point.

  6. Test it for real. Type shiny_app(launch_browser = TRUE) and press Enter. This will run the Shiny app. You may get an error message like this:

    Loading required package: shiny

Listening on http://127.0.0.1:3758
Warning: Error in loadNamespace: namespace ‘rlang’ 0.4.12 is already loaded, but >= 1.0.0 is required
  50: list2
  49: shiny::reactiveValues
  48: server
Error in loadNamespace(i, c(lib.loc, .libPaths()), versionCheck = vI[[i]]) : 
  namespace ‘rlang’ 0.4.12 is already loaded, but >= 1.0.0 is required

    If so, quit the Shiny app (not RStudio!; press escape one or two times in the Console) and run the command install.packages("rlang"). You may have to repeat for other packages that are used by growthis.

  7. You can quit the app by pressing the Escape key in the Console.

Usage of the growthis Shiny app

These steps will need to be repeated every time you start RStudio to work with the app. In the Console panel, type these commands.

  1. Type library(growthis) and press Enter.
  2. Type shiny_app(launch_browser = TRUE) and press Enter.

Alternatively, you can start the app in one command: growthis::shiny_app(launch_browser = TRUE) and press Enter. You can also run the command shiny_app() and then open the URL (adress) which is reported in the console manually (http://127.0.0.1:3838).

Quit the app by pressing the Escape key in the Console. The browser tab should be closed as well. It is not active anymore. Only closing the browser tab will not shut down the app. Note that an (inactive) running app in RStudio may drain the battery of your laptop.

Updating the growthis package

The package is in development and releases follow each other frequently. You can update the current package by typing the following in the Console:

  1. Type library(devtools) and press Enter.
  2. Type install_github("MichielNoback/growthis") and press Enter.

This will update the package to the most recent version.

Usage of the other growthis package functions.

If you are prepared to use R, several functions are at your disposal. Type library(growthis) to get it in the current session. Type growthis:: in the console and you will see all available datasets and functions. All functions have within-R documentation. You can see these in the Help panel when you type ?functions_name in the Console (in RStudio of course).
The main functions are:

  • read_experiment_data() to load from Excel files (this requires some metadata to be present).
  • filter_data() to filter imported datasets.
  • plot_growthcurves() to make plots.
  • do_growth_analysis() will give growth statistics. This requires the input of data in "wide" format; the function create_wide_data() does exactly that.

In the console, type ?<function name> to get details on their usage.