Repository files navigation

FRApp is an R package and an R-based Shiny application providing auser-friendly interactive interface to streamline the data analysisderived by fitting nonlinear mixed-effects regression models with anasymptotic exponential functional relationship on data with ahierarchical structure. The package provides a comprehensive suite oftools tailored for efficient data analysis and visualization.

The package includes data from Fluorescence Recovery AfterPhotobleaching (FRAP) experiments on actin dynamics in dendritic spines.

To explore all the functionalities ofFRApp, a comprehensive guideis available in a dedicated vignette.

To use theFRApp application first download and install R andRStudio fromposit.co.

You can install the package from the CRAN executing the followingcommands from the RStudio console:

install.packages("FRApp")

You can install the development version ofFRApp fromGitHub with:

# install.packages("devtools")devtools::install_github("gioiadc/FRApp")

To load the package and make the functions available to be used run thecommand:

library(FRApp)

You must execute this command every time you restart RStudio.

To launch the Shiny app, copy and paste the following code in theconsole:

FRApp()

The application opens automatically in the browser.

FRApp allows you to:

• load the data

• estimate and compare exponential mixed-effects models

• print a model report

• export the data and the model corresponding objects.

TheFit the model button allows you to estimate the model. At theend of the estimation process, the following objects will appear on theright side of the application:

• summary information on the model

• diagnostic plots: scatterplot of residuals vs. estimated values andquantile-quantile plot of the residuals

• approximate 95% confidence intervals of the parameters of interest.

Model fitting is performed with the functionnlme.

You can save models for model comparison with the buttonAdd to modellist in theModel list section. Saved models must have differentnames. When multiple models are saved, the table at the bottom of thepage displays comparisons between the saved models using AIC, BIC, andlikelihood ratio test criteria.

TheReset model list button allows you to delete the saved modelsfrom the list.

From the drop-down menu in theDownload section, it is possible toselect a model saved in the model list and download a report and somemodel’s objects.

In theDownload section, theReport button allows you to exportthe results printed in the application (model summary, residual graphs,and intervals) into a PDF document.

TheRData button allows you to export an RData file containing sixobjects:

• data: the dataset in the format used for the analysis

• fit: the output of the estimated model

• pred: the values estimated by the model

• CI: the approximate 95% confidence intervals

• resid: the residuals of the model

• raneff: the random effects available at the different hierarchicallevels.

As a reference for the construction of the model and the differentoptions to specify, we refer to the book: Pinheiro, J., & Bates, D.(2006). Mixed-effects models in S and S-PLUS. Springer science &business media.

As a reference for the FRAP data analysis example: Di Credico, G.,Pelucchi, S., Pauli, F. et al. Nonlinear mixed-effects models to analyzeactin dynamics in dendritic spines.Scientific Reports 15, 5790(2025).https://doi.org/10.1038/s41598-025-87154-w