FitGrowth is a Shiny app for (bacterial) growth curve analysis I wrote recently. You can get it on GitHub, where you will find also example files for testing. For a quick vew, the app is also running on the amazon cloud here.

With FitGrowth you can:

  • Upload your own growth data.
  • Visualize it and fit a model on it.
  • Select interactively the time interval for fitting.
  • Select number of facets of the plot and which samples to analyze.
  • Obtain the model parameters as a table.
  • Analyse many samples.
  • Download all plots and tables.

fitgrowth1.png fitgrowth2.png

How it works

This app fits growth data to the continuous logistic equation. The best parameters n0, k and r are found using the nonlinear least-squares method nls in R. The app handles one or many samples (tested with 96), as well as NA values. You can get an example file here. The easiest way to run FitGrowth locally is if you have RStudio, you just have to paste this in your console:

shiny::runGitHub('FitGrowth', 'angelovangel')

You will need these packages, so install them if you haven’t done so: shiny, shinydashboard, modelr, tidyverse, broom and DT.

Instructions for use

Load the data as a text file, the first column must be named time, all other columns are treated as samples and the columns names are used as sample names. Adjust the file input settings until the data is read into the app. After that, take a look at the other tabs. Note that the parameters of the logistic model are re-calculated when the time interval is changed with the slider.

Feedback and suggestions are welcome!

Cheers!

Angel