Skip to content

bharathananth/compareRhythms

Repository files navigation

compareRhythms

License: GPL-3 DOI

The goal of compareRhythms is to find features with altered circadian rhythm parameters (amplitude and phase) between the control and experimental groups. A fit of the function $A\cos(2\pi t/24 - \phi)$ to the data defines the rhythm parameters amplitude ($A$) and phase ($\phi$).

Installation

You can install the current version of compareRhythms from GitHub with:

if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
BiocManager::install(version = "3.19")   # This installs bioconductor, but version 3.15 onwards are also ok
BiocManager::install(c("SummarizedExperiment", "DESeq2", "edgeR", "limma", "rain")) # Packages needed by compareRhythms
install.packages("devtools")    # if it is not already installed
devtools::install_github("cran/npsm")   # Package archived by CRAN
devtools::install_github("cran/DODR")   # Package archived by CRAN
devtools::install_github("bharathananth/compareRhythms", build_vignettes = TRUE, dependencies = TRUE)

Usage summary

The analysis is run using a single function compareRhythms(). To execute this function, the three necessary ingredients are the timeseries data, the experimental design and parameters to choose and tune the method. The output of the function is a data.frame with the IDs of the differentially rhythmic features, the category it is classified under and optionally the rhythm parameters of the feature in the two groups. The differential rhythmicity categories are gain of, loss of, change of, or same rhythms (with respect to the reference/control group).

For complete examples, please refer to the vignette titled User Guide by running

library(compareRhythms)
vignette("UserGuide", package="compareRhythms")

or view the documentation.

Citation

Please cite compareRhythms in publications as:

The R code to perform all the analyses in this publication (using compareRhythms) can be found in the repository https://github.com/bharathananth/FEBSJ-VDA-overestimates

Getting help

If you encounter a bug, please file a minimal reproducible example on github.