Learn R Programming
⚠️There's a newer version (0.1.2) of this package.Take me there.

cobiclust

cobiclust is an R package dedicated to the biclustering of over-dispersed count data such as those produced by amplicon-based sequencing for example.

How to install cobiclust?

To install the cobiclust package within R from GitHub, open a R session and:

  • install devtools with install.packages("devtools") (if not installed yet)
  • load the devtools R package with library(devtools)
  • run install_github("julieaubert/cobiclust")

About cobiclust

The cobiclust package has been developped at UMR518 MIA-Paris (INRA, AgroParisTech) by J. Aubert (julie.aubert@agroparistech.fr).

Copy Link

Version

Install

install.packages('cobiclust')

Monthly Downloads

280

Version

0.1.0

License

GPL-2

Maintainer

J Aubert

Last Published

October 15th, 2018

Functions in cobiclust (0.1.0)

penalty

Calculate the BIC penalty
selection_criteria

Calculate selection criteria.
qukg_calculation

Calculate E_q(U_ijr) and E_q(log U_ijr) when U_ijr follows a Gamma(a_k,g,a_k,g)
foo_a

Useful function to estimate the parameter a
init_pam

Initialisation of the co-clusters by partitioning around medoids method.
is.cobiclustering

Is an object of class cobiclustering ?
lb_calculation

Calculate the lower bound
alpha_calculation

Calculate the matrix (alpha_k,g)
cobiclust

Perform a biclustering adapted to overdispersed count data.
cobiclustering

Creation of the cobiclustering class.
dicho

Internal function - estimation of the parameter a
qu_calculation

Calculate E_q(U_ijr) and E_q(log U_ijr) when U_ijr follows a Gamma(a,a)
summary.cobiclustering

Summary of an object of class Cobiclust