CRAN Task View: gRaphical Models in R

Wikipedia defines a graphical model as a graph that represents independencies among random variables by a graph in which each node is a random variable, and the missing edges between the nodes represent conditional independencies . This task view is a collection of packages intended to supply R code to deal with graphical models.

Note: I have recently become maintainer of the gR task view and I am currently in the process of updating the view. Please do contact me if a package is missing, a description is inadequate etc.

The packages can be roughly structured into the following topics (although several of them have functionalities which go across these categories):

Representation, manipulation and display of graphs

• graph [vignette: y; demo: n] A package to handle graph data structures.
• RBGL [vignette: y; demo: n] Interface to boost C++ graph library (based on graph objects from the graph package).
• gRain [vignette: y; demo: y] An alternative interface to graphs. Uses the graph and RBGL and implements additional graph operations.
• dynamicGraph [vignette: n; demo: y] Provides an advanced graphical user interface for display and interaction with graphs (based on Tcl/Tk).
• mathgraph [vignette: n; demo: n] Implements matrix representations of graphs and provide a plot function.
• diagram [vignette: n; demo: y] Functions for visualising simple graphs (networks), plotting flow diagrams
• giRaph [vignette: n; demo: y] Supply classes and methods to represent and manipulate graphs
• igraph [vignette: n; demo: n] Routines for simple graphs, network analysis.

Classical models - General purpose packages

• gRbase [vignette: y; demo: y] Sets up a meta data representation which is used by several other packages (among these are mimR and gRain). gRbase suggests a way to implement models and fitting engines - illustrated by hierarchical log-linear models.
• mimR [vignette: n; demo: n] General package for model selection in contingency tabels, graphical Gaussian models and mixed interaction models. The package provides an interface to the MIM program and only runs on Windows.
• CoCo [vignette: n; demo: y] An advanced suite of functions for contingency tabels, graphical Gaussian models and mixed interaction models.
• ggm [vignette: n; demo: n] Fitting graphical Gaussian models.

Miscellaneous: Model search, specialized types of models etc.

• SIN [vignette: n; demo: n] SIN model selection in Gaussian graphical Markov models.
• pcalg [vignette: y; demo: n] Estimating the skeleton and equivalence class of a DAG
• qp [vignette: n; demo: n] q-order partial correlation graph search algorithm. A robust procedure for structure learning of undirected Gaussian graphical Markov models from "small n, large p" data.
• gRc [vignette: n; demo: n] Inference in graphical Gaussian models with edge and vertex symmetries (Graphical Gaussian models with colours).
• rggm [vignette: n; demo: n] Robustified methods for Gaussian Graphical Models.
• GeneNet [vignette: n; demo: n] Modeling and Inferring Gene Networks. GeneNet is a package for analyzing gene expression (time series) data with focus on the inference of gene networks.

Bayesian Networks/Probabilistic expert systems

• gRain [vignette: y; demo: n] Implements propagation (message passing) in graphical models (Bayesian networks).
• deal [vignette: n; demo: n] Learning Bayesian networks with mixed (discrete and continuous) variables.

BUGS models

• R2WinBUGS [vignette: y; demo: n] Running WinBUGS and OpenBUGS from R / S-PLUS
• rbugs [vignette: n; demo: n] Fusing R and OpenBugs
• BRugs [vignette: n; demo: n] Provides an R interface on Windows machines to OpenBUGS. It works only under Windows and used to be available from CRAN, now it is located at the CRANextras repository.
• boa [vignette: n; demo: n] Bayesian Output Analysis Program (BOA) for MCMC.
• coda [vignette: n; demo: n] Output analysis and diagnostics for MCMC.
• bayesmix [vignette: n; demo: n] Bayesian network structure learning
• bnlearn [vignette: n; demo: n] Bayesian network structure learning
• G1DBN [vignette: n; demo: n] A package performing Dynamic Bayesian Network inference.
• ergm [vignette: n; demo: n] Fit, Simulate and Diagnose Exponential-Family Models for Networks

CRAN packages:

• bayesmix
• bnlearn
• boa
• CoCo
• coda
• deal
• diagram
• dynamicGraph
• ergm
• G1DBN
• ggm
• giRaph
• gRain
• graph
• gRbase (core)
• gRc
• igraph
• mathgraph
• mimR
• pcalg
• qp
• R2WinBUGS
• RBGL
• rbugs
• rggm
• SIN

Related links:

• gR homepage and mailing list
• Bioconductor Package: Rgraphviz
• Bioconductor
• JAGS
• OpenBUGS
• MIM
• Bayes Net Toolbox (for MATLAB)
• DIGRAM
• TETRAD
• Grappa