iCARH: Integrative Conditional Autoregressive Horseshoe Model

Implements the integrative conditional autoregressive horseshoe model discussed in Jendoubi, T., & Ebbels, T. (2018). "Integrative analysis of time course metabolic data and biomarker discovery". arXiv preprint <arXiv:1801.07767>. The model consists in three levels: Metabolic pathways level modeling interdependencies between variables via a conditional auto-regressive (CAR) component , integrative analysis level to identify potential associations between heterogeneous omic variables via a Horseshoe prior and experimental design level to capture experimental design conditions through a mixed-effects model. The package also provides functions to simulate data from the model, construct pathway matrices, post process and plot model parameters.

Depends: rstan, MASS, stats, ggplot2
Imports: RCurl, KEGGgraph, igraph, reshape2, mc2d, abind, Matrix
Suggests: knitr, rmarkdown
Published: 2020-01-14
Author: Takoua Jendoubi [aut, cre], Timothy M.D. Ebbels [aut]
Maintainer: Takoua Jendoubi <t.jendoubi14 at imperial.ac.uk>
License: GPL (≥ 3)
NeedsCompilation: no
CRAN checks: iCARH results


Reference manual: iCARH.pdf
Vignettes: Example of simulating and running the iCARH model
Package source: iCARH_2.0.1.1.tar.gz
Windows binaries: r-prerelease: iCARH_2.0.1.1.zip, r-release: iCARH_2.0.1.1.zip, r-oldrel: iCARH_2.0.1.1.zip
macOS binaries: r-prerelease: not available, r-release: iCARH_2.0.1.1.tgz, r-oldrel: iCARH_2.0.1.1.tgz
Old sources: iCARH archive


Please use the canonical form https://CRAN.R-project.org/package=iCARH to link to this page.