This repository is for analyising a WT mice dataset to test for sex differences in mean and variance for over 200 traits. Our intial questions were two-fold:
1) Do males and females show consistent sex differences / biases in traits? 2) Do males and females show similar or different biases in trait variance?
The html file, published here https://rpubs.com/SusZaj/ESF, or download (and "save file") to open locally) provides the most comprehensive overview of our workflow and analyses without having to rerun the analyses yourself.
The shinyApp to look at male / female bias in means and variablity in traits, procedures and functional groups can be found here: https://szaj.shinyapps.io/SexDifference_Shiny/
Unfortunately, the raw data file can not be provided via Github, as it is too large (274MB). However, it is freely accessible and uploaded to zenodo.org. As such, we have already processed the raw data file and provide a cleaned up file which is less computationally intensive to deal with. The file has been saved in a folder called export. This file is used for all further processing and analysis.
If users do not already have R and RStudio they will need to download both from the CRAN mirrors page. Choose the mirror closest to your locatioon and it will link to the R download page. Click the version and operating system that is relevant to you. RStudio can be downloaded from the RStudio download page.
To re-run the analysis users can clone or download the entire respository to their local machine (un-zip the file if downloaded). Click on the
mice_sex_diff.Rproj. This will open up RStudio and set the working directory to the respository.
Once Rstudio is open, navigate the MouseSexDiffVar_Oct_2020.Rmd file in the scripts/ folder. This file has all the relevant code for analyses and provides detailed annotation on how things were done. One can simply walk through the code chunks as they appear, or more easily, just knit the entire document to html using knit to html_document2 in the tab knit. Note that, to do this, you need to install the bookdown package from CRAN using install.packages("bookdown"). Other packages necessary for the scrpt will be loaded using pacman.
If user have any questions or issues in running or understanding the code please contact the authors: Susi Zajitschek or Shinichi Nakagawa