Applications/PaCeQuantAna

From MiToBo
Revision as of 18:10, 24 July 2020 by Poeschl (talk | contribs)
Jump to navigationJump to search

PaCeQuantAna: R package for Visualization and Statistical Analysis of Cell Shape Featues

We provide an R package for visualizing and analyzing features extracted by PaCeQuant.


The new version compatible with PaCeQuant as released in MiToBo 2.0 in May 2020 is out now! It can be downloaded here:


Installation

For installation in RStudio download the package and follow the instructions below:

  1. Install the following packages:
    • caroline
    • gplots
    • dunn.test
    • sm
    • RColorBrewer

    To this end start RStudio, select

    “Tools” → “Install Packages…”

    from the menu bar.
    Choose

    “Install from: Repository (CRAN)”

    and enter the names of the packages.
    Activate

    “Install dependencies”.

  2. The multtest package can be installed via the following Bioconductor installation script:

    source ("http://bioconductor.org/biocLite.R")

    biocLite("multtest")

  3. Install the PaCeQuantAna package selecting

    “Tools” → “Install Packages…”

    from the menu bar.

    Choose

    “Install from: Package Archive File (.tgz; .tar.gz)"

    and select the tar.gz file of PaCeQuantAna via

    "Browse...".


We use Arial or Helvetica fonts when creating the plots, which is often recommended for publication. Therefore you will need to prepare your working environment by once running

library(extrafont)

font_import()

loadfonts()

This will initialize the creation of a font database.
You can check which fonts where found and load by

fonts()

Documentation

Find more details of how to use the package and details about the general workflow in the vignette of the package by either navigating to the vignette in the Rstudio package manager tab or by typing

vignette("PaCeQuantAna")

in the console.

Please note:
The vignette is currently under construction. An updated version will be released soon.

The vignette contains comprehensive documentation of the various functions and options of the package.
In addition, a sample workflow R script is provided which shows all commands to load your data, configure the output and do the analysis.
You can directly copy the script into a file and load this file into the Rstudio editor to run it completely or line-wise.

You can test the new version of the package with sample data and small workflows for analyses of developmental stages and mutant screening. Therefore we provide two archives:

Both archives contain the specific

  • sample data
  • workflows and
  • data description files

. The sample data sets are reduced to 2 microscopy images per group (time point or species). The workflows contain the function calls depicted in Fig. 5 in Poeschl et al 2020. The data description files are customized to fit e.g. the species and time points included in our analyses, color used or font face used.

Both archives include ready to use workflows for provided sample data sets. Only the path to the folder of the extracted archive needs to be set as working directory in the corresponding R-scripts.
In detail developmentalStages.zip contains

  • sample data: Time-series_2-3-5-7, a folder containing the data organised into folders
  • 3 workflows: 1 general workflow for analyses of developmental stages and 2 specific workflows for
    • Col-0: including prepared outputFolder containing the customised data description file
    • Col-0_iqd5-1_iqd5-2: including prepared outputFolder containing the customised data description file


In detail mutantScreening.zip contains

  • sample data: Mutants, a folder containing the data organised into folders
  • 2 workflows: 1 general workflow for screening mutants and 1 specific workflow for
    • mutant_screening: including prepared outputFolder containing the customised data description file


Important notes

  • By default the working directory (set with setwd(...)) where the output folder will be created, and the data directory (set with dataDir) must not be identical!
    If you require these folders to be identical for whatever reason you need to manually edit the file data_description.csv in the output folder and delete the row referring to the output folder.
  • On Windows operating systems you may encounter problems with missing fonts, i.e., error messages like the following one:

    ...

    Error in pdf(file = paste0(outdir, fsep, base_file_name,

    "_dunn_test_heatmap_adjusted_pvalues.pdf"),  :

      unknown family 'Arial'

    ...

    Initialize each of your workflows using PaCeQuantAna with

    library(extrafont)

    This will initialize the Windows font database and should solve the problems.



Email
Email.png

Don't hesitate to send an email when you encounter a problem or when you have questions: poeschl@informatik.uni-halle.de


Old versions

The old version of PaCeQuantAna compatible with PaCeQuant results generated with MiToBo 1.8.x can be downloaded here:

You can test the package on these sample data: sample data for PaCeQuantAna