Merge branch 'dev'

DESCRIPTION

@@ -1,26 +1,27 @@
 Package: sparsebn
 Title: Learning Sparse Bayesian Networks from High-Dimensional Data
 Version: 0.0.5
-Date: 2017-03-15
-Authors@R: person("Bryon", "Aragam", email = "sparsebn@gmail.com", role = c("aut", "cre"))
+Date: 2017-09-11
+Authors@R: c(
+    person("Bryon", "Aragam", email = "sparsebn@gmail.com", role = c("aut", "cre")),
+    person("Jiaying", "Gu", role = c("aut")),
+    person("Dacheng", "Zhang", role = c("aut")),
+    person("Qing", "Zhou", role = c("aut"))
+    )
 Maintainer: Bryon Aragam <sparsebn@gmail.com>
-Description: Fast methods for learning sparse Bayesian networks from high-dimensional data using sparse regularization, as described in Aragam, Gu, and Zhou (2017) <https://arxiv.org/abs/1703.04025>. Designed to handle mixed experimental and observational data with thousands of variables with either continuous or discrete observations.
+Description: Fast methods for learning sparse Bayesian networks from high-dimensional data using sparse regularization, as described in Aragam, Gu, and Zhou (2017) <arXiv:1703.04025>. Designed to handle mixed experimental and observational data with thousands of variables with either continuous or discrete observations.
 Depends:
     R (>= 3.2.3),
     sparsebnUtils (>= 0.0.5),
     ccdrAlgorithm (>= 0.0.4),
-    discretecdAlgorithm (>= 0.0.3)
+    discretecdAlgorithm (>= 0.0.5)
 Suggests:
     knitr,
     rmarkdown,
     mvtnorm,
     igraph,
     graph,
     testthat
-Remotes:
-    itsrainingdata/sparsebnUtils@dev,
-    itsrainingdata/ccdrAlgorithm@dev,
-    itsrainingdata/discretecdAlgorithm
 URL: https://github.com/itsrainingdata/sparsebn
 BugReports: https://github.com/itsrainingdata/sparsebn/issues
 License: GPL (>= 2)

NEWS.md

@@ -1,3 +1,12 @@
+# sparsebn 0.0.5
+
+## Features
+
+* `estimate.dag` now supports white lists and black lists (#6)
+* Cytoscape compatibility now available via the method `openCytoscape` (#4)
+* `plotDAG` now includes labels for each subplot by default
+* See NEWS files for sparsebnUtils, discretecdAlgorithm, and ccdrAlgorithm for more updates
+
 # sparsebn 0.0.4
 
 ## Notes

R/data.R

@@ -1,10 +1,13 @@
 #' The pathfinder network
 #'
-#' Simulated data and network for the pathfinder network from the \href{http://www.bnlearn.com/bnrepository/#pathfinder}{Bayesian network repository}.
-#' Pathfinder is an expert system developed by Heckerman et. al (1992)
+#' Simulated data and network for the pathfinder network from the
+#' \href{http://www.bnlearn.com/bnrepository/#pathfinder}{Bayesian network repository}.
+#' Pathfinder is an expert system developed by
+#' \href{http://heckerman.com/david/HN92cbr.pdf}{Heckerman et. al (1992)} [1]
 #' to assist with the diagnosis of lymph-node diseases.
 #'
-#' The data is simulated from a Gaussian SEM assuming unit edge weights and
+#' This is a benchmark network used to test algorithms for learning Bayesian
+#' networks. The data is simulated from a Gaussian SEM assuming unit edge weights and
 #' unit variances for all nodes.
 #'
 #' @format A \code{\link{list}} with four components:
@@ -22,28 +25,36 @@
 #' @usage
 #' data(pathfinder)
 #'
+#' @references
+#' [1] Heckerman, David E., and Bharat N. Nathwani. "\href{http://heckerman.com/david/HN92cbr.pdf}{An evaluation of the diagnostic accuracy of Pathfinder}." Computers and Biomedical Research 25.1 (1992): 56-74.
+#'
 #' @examples
-#' \dontrun{
-#' # Create a valid sparsebnData object from the simulated pathfinder data
+#'
+#' ### Create a valid sparsebnData object from the simulated pathfinder data
 #' data(pathfinder)
 #' dat <- sparsebnData(pathfinder$data, type = "c")
 #'
-#' # If desired, change the edge weights to be randomly generated
-#' coefs <- get.adjacency.matrix(pathfinder$dag)
-#' coefs[coefs != 0] <- runif(n = num.edges(pathfinderDAG), min = 0.5, max = 2)
-#' vars <- Matrix::Diagonal(n = num.nodes(pathfinderDAG), x = rep(1, num.nodes(pathfinderDAG)))
-#' id <- vars
-#' covMat <- t(solve(id - coefs)) %*% vars %*% solve(id - coefs)
-#' pathfinder.data <- rmvnorm(n = 1000, sigma = as.matrix(covMat))
-#' }
+#' ### Code to reproduce this dataset by randomly generating edge weights
+#' coefs <- runif(n = num.edges(pathfinder$dag), min = 0.5, max = 2) # coefficients
+#' vars <- rep(1, num.nodes(pathfinder$dag))                         # variances
+#' params <- c(coefs, vars)                                          # parameter vector
+#' pathfinder.data <- generate_mvn_data(graph = pathfinder$dag,
+#'                                      params = params,
+#'                                      n = 1000)
 #'
 "pathfinder"
 
 #' The discrete cytometry network
 #'
 #' Data and network for analyzing the flow cytometry experiment
-#' from \href{http://science.sciencemag.org/content/308/5721/523.long}{Sachs et al. (2005)}.
-#' The data is a cleaned and discretized version of the raw data (see \code{\link{cytometryContinuous}}) from these experiments.
+#' from \href{http://science.sciencemag.org/content/308/5721/523.long}{Sachs et al. (2005)} [1].
+#' The data is a cleaned and discretized version of the raw data (see
+#' \code{\link{cytometryContinuous}} for details) from these experiments.
+#'
+#' After cleaning and pre-processing, the raw continuous measurements have been
+#' binned into one of three levels: low = 0, medium = 1, or high = 2. Due to the
+#' pre-processing, the discrete data contains fewer observations (n = 5400)
+#' compared to the raw, continuous data.
 #'
 #' @format A \code{\link{list}} with three components:
 #'
@@ -57,6 +68,9 @@
 #' @usage
 #' data(cytometryDiscrete)
 #'
+#' @references
+#' [1] Sachs, Karen, et al. "\href{http://science.sciencemag.org/content/308/5721/523.long}{Causal protein-signaling networks derived from multiparameter single-cell data}." Science 308.5721 (2005): 523-529.
+#'
 #' @examples
 #' # Create a valid sparsebnData object from the cytometry data
 #' data(cytometryDiscrete)
@@ -67,9 +81,16 @@
 #' The continuous cytometry network
 #'
 #' Data and network for analyzing the flow cytometry experiment
-#' from \href{http://science.sciencemag.org/content/308/5721/523.long}{Sachs et al. (2005)}.
+#' from \href{http://science.sciencemag.org/content/308/5721/523.long}{Sachs et al. (2005)} [1].
 #' This dataset contains the raw measurements from these experiments.
 #'
+#' The dataset consists of n = 7466 observations of p = 11 continuous
+#' variables corresponding to different proteins and phospholipids in human
+#' immune system cells, and each observation indicates the measured level of
+#' each biomolecule in a single cell under different experimental interventions.
+#' Based on this data, a consensus network was reconstructed and validated, which
+#' is included as well.
+#'
 #' @format A \code{\link{list}} with three components:
 #'
 #' \itemize{
@@ -82,6 +103,9 @@
 #' @usage
 #' data(cytometryContinuous)
 #'
+#' @references
+#' [1] Sachs, Karen, et al. "\href{http://science.sciencemag.org/content/308/5721/523.long}{Causal protein-signaling networks derived from multiparameter single-cell data}." Science 308.5721 (2005): 523-529.
+#'
 #' @examples
 #' # Create a valid sparsebnData object from the cytometry data
 #' data(cytometryContinuous)

R/sparsebn-main.R

@@ -29,6 +29,14 @@
 #'                used based on a decreasing log-scale  (see also \link[sparsebnUtils]{generate.lambdas}).
 #' @param lambdas.length Integer number of values to include in the solution path. If \code{lambdas}
 #'                       has also been specified, this value will be ignored.
+#' @param whitelist A two-column matrix of edges that are guaranteed to be in each
+#'                  estimate (a "white list"). Each row in this matrix corresponds
+#'                  to an edge that is to be whitelisted. These edges can be
+#'                  specified by node name (as a \code{character} matrix), or by
+#'                  index (as a \code{numeric} matrix).
+#' @param blacklist A two-column matrix of edges that are guaranteed to be absent
+#'                  from each estimate (a "black list"). See argument
+#'                  "\code{whitelist}" above for more details.
 #' @param error.tol Error tolerance for the algorithm, used to test for convergence.
 #' @param max.iters Maximum number of iterations for each internal sweep.
 #' @param edge.threshold Threshold parameter used to terminate the algorithm whenever the number of edges in the
@@ -56,6 +64,8 @@
 estimate.dag <- function(data,
                          lambdas = NULL,
                          lambdas.length = 20,
+                         whitelist = NULL,
+                         blacklist = NULL,
                          error.tol = 1e-4,
                          max.iters = NULL,
                          edge.threshold = NULL,
@@ -89,11 +99,22 @@ estimate.dag <- function(data,
     ### Is the data gaussian, binomial, or multinomial? (Other data not supported yet.)
     data_family <- sparsebnUtils::pick_family(data)
 
+    ### If intervention list contains character names, convert to indices
+    if("character" %in% sparsebnUtils::list_classes(data$ivn)){
+        data$ivn <- lapply(data$ivn, function(x){
+            idx <- match(x, names(data$data))
+            if(length(idx) == 0) NULL # return NULL if no match (=> observational)
+            else idx
+        })
+    }
+
     ### Run the main algorithms
     if(data_family == "gaussian"){
         ccdrAlgorithm::ccdr.run(data = data,
                                 lambdas = lambdas,
                                 lambdas.length = lambdas.length,
+                                whitelist = whitelist,
+                                blacklist = blacklist,
                                 gamma = concavity,
                                 error.tol = error.tol,
                                 max.iters = max.iters,
@@ -104,6 +125,8 @@ estimate.dag <- function(data,
         discretecdAlgorithm::cd.run(indata = data,
                                     lambdas = lambdas,
                                     lambdas.length = lambdas.length,
+                                    whitelist = whitelist,
+                                    blacklist = blacklist,
                                     error.tol = error.tol,
                                     convLb = convLb,
                                     weight.scale = weight.scale,

R/sparsebn-plotting.R

@@ -72,6 +72,7 @@ plotDAG.sparsebnPath <- function(x, ...){
     #
 
     plot(x,
+         labels = TRUE,
          vertex.size = 4,
          vertex.label = NA,
          vertex.label.color = gray(0),

README.Rmd

@@ -14,6 +14,7 @@ knitr::opts_chunk$set(
 
 # sparsebn
 
+[![Project Status: Active  The project has reached a stable, usable state and is being actively developed.](http://www.repostatus.org/badges/latest/active.svg)](http://www.repostatus.org/#active)
 [![Travis-CI Build Status](https://travis-ci.org/itsrainingdata/sparsebn.svg?branch=master)](https://travis-ci.org/itsrainingdata/sparsebn)
 [![](http://www.r-pkg.org/badges/version/sparsebn)](http://www.r-pkg.org/pkg/sparsebn)
 [![CRAN RStudio mirror downloads](http://cranlogs.r-pkg.org/badges/sparsebn)](http://www.r-pkg.org/pkg/sparsebn)

README.md

@@ -1,7 +1,7 @@
 sparsebn
 ========
 
-[![Travis-CI Build Status](https://travis-ci.org/itsrainingdata/sparsebn.svg?branch=master)](https://travis-ci.org/itsrainingdata/sparsebn) [![](http://www.r-pkg.org/badges/version/sparsebn)](http://www.r-pkg.org/pkg/sparsebn) [![CRAN RStudio mirror downloads](http://cranlogs.r-pkg.org/badges/sparsebn)](http://www.r-pkg.org/pkg/sparsebn)
+[![Project Status: Active The project has reached a stable, usable state and is being actively developed.](http://www.repostatus.org/badges/latest/active.svg)](http://www.repostatus.org/#active) [![Travis-CI Build Status](https://travis-ci.org/itsrainingdata/sparsebn.svg?branch=master)](https://travis-ci.org/itsrainingdata/sparsebn) [![](http://www.r-pkg.org/badges/version/sparsebn)](http://www.r-pkg.org/pkg/sparsebn) [![CRAN RStudio mirror downloads](http://cranlogs.r-pkg.org/badges/sparsebn)](http://www.r-pkg.org/pkg/sparsebn)
 
 Introducing `sparsebn`: A new R package for learning sparse Bayesian networks and other graphical models from high-dimensional data via sparse regularization. Designed from the ground up to handle:
 

cran-comments.md

@@ -1,20 +1,23 @@
 ## Test environments
-* local OS X install, R 3.3.3
-* ubuntu 12.04.5 (travis-ci), R 3.3.3 (oldrel, devel, and release)
+* local OS X install, R 3.4.1
+* ubuntu 12.04.5 (travis-ci: oldrel, devel, and release)
 * win-builder (devel and release)
-* r-hub (oldrel, devel, and release)
+* r-hub (devel)
 
 ## R CMD check results
-There were no ERRORs or WARNINGs. There was one NOTE:
+There were no ERRORs, WARNINGs, or NOTEs.
 
-* checking CRAN incoming feasibility ... NOTE
-Maintainer: ‘Bryon Aragam <sparsebn@gmail.com>’
+## CRAN Package Check Results for Package sparsebn 
 
-Days since last update: 4
+From https://cran.rstudio.com/web/checks/check_results_sparsebn.html
 
-- This is a very minor release to update the package metadata to include a reference
-to a new preprint discussing this package. No DOI is available yet since it is currently
-under review.
+Version: 0.0.4
+Check: package dependencies
+Result: NOTE
+    Package suggested but not available for checking: ‘graph’
+Flavor: r-release-osx-x86_64
+
+This has been fixed.
 
 ## Dependencies