New script

ABC-Install_all_packages.R

@@ -0,0 +1,161 @@
+# tocID <- "ABC-Install_all_packages.R"
+#
+# Purpose:  A Bioinformatics Course:
+#              Installing all packages in this course
+#
+# Version:  1.0
+#
+# Date:     2021  10
+# Author:   Boris Steipe (boris.steipe@utoronto.ca)
+#
+# Versions:
+#           1.0    New code
+#
+#
+# TODO:
+#
+# ==============================================================================
+
+
+#TOC> ==========================================================================
+#TOC>
+#TOC>   Section  Title                          Line
+#TOC> ----------------------------------------------
+#TOC>   1        Packages                         33
+#TOC>   2        CRAN packages                    98
+#TOC>   3        Bioconductor packages           127
+#TOC>   4        Other package sources           142
+#TOC>   5        Updating packages               148
+#TOC>
+#TOC> ==========================================================================
+
+
+# =    1  Packages  ============================================================
+
+# Much of R's functionality is contributed in packages: bundles of R scripts
+# or code in other languages, pre-configured objects, and datasets. Making this
+# functionality available is often done by issuing a library(<package-name>)
+# command, however this is not the preferred way, since it may override other
+# R functions and it makes it harder to understand where the source code of
+# a particular function is located. In this course we call the function name
+# prefixed with the package name and two colons:
+#   <package-name>::<function-name>()
+# This is the preferred way, since it is explicit.
+#
+# Regardless of which idiom one uses to call the actual function, the package
+#  needs to be "installed" first, i.e. the code must have been downloaded
+# from CRAN, or using the BiocManager::install() function.
+#
+# This script contains download commands for all packages that are used in the
+# course. You can execute the script line by line (or even source the entire
+# script) to make sure all packages can be installed on your computer. Just
+# one reminder: if you are ever asked to install from source, the correct
+# answer is usually "no" - except if you really know what you are doing and why.
+#
+# Once packages are installed you can get additional information about
+# the contents of a package with the commands:
+#  library(help=<package-name>)       # basic information
+#  browseVignettes("<package-name>")  # available vignettes
+#  data(package = "<package-name>")   # available datasets
+#
+#  ... and you can load data sets with:
+#  data(<data-set-name>, package = "<package-name>")
+#
+#  All packages here are installed only when they have not been installed
+#  before, using the following idiom:
+#
+#     if (! requireNamespace("<package-name>", quietly=TRUE)) {
+#       install.packages("<package-name>")
+#     }
+#
+#  ... or its BiocManager::install() equivalent:
+#
+# if (! requireNamespace("<bioconductor-package-name>", quietly=TRUE)) {
+#   BiocManager::install("<bioconductor-package-name>")
+# }
+#
+#  If you want to _force_ a re-installation of the package, simply issue
+#  the install.packages("<package-name>") command on its own. For compactness
+#  we wrap the idiom into a function, which can also switch between CRAN
+#  and BIOconductor sources:
+
+installIfNeeded <- function(package, s = "CRAN") {
+  # s: "CRAN" or "BIO"
+  if (s == "CRAN") {
+    if (! requireNamespace(package, quietly=TRUE)) {
+      install.packages(package)
+    }
+  } else if (s == "BIO") {
+    if (! requireNamespace("BiocManager", quietly=TRUE)) {
+      install.packages("BiocManager")
+    }
+    if (! requireNamespace(package, quietly=TRUE)) {
+      BiocManager::install(package)
+    }
+  } else {
+    stop(sprintf("Unknown source \"%s\".", s))
+  }
+}
+
+
+# =    2  CRAN packages  =======================================================
+
+installIfNeeded("ape")
+installIfNeeded("BiocManager")
+installIfNeeded("bio3d")
+installIfNeeded("evd")
+installIfNeeded("ggseqlogo")
+installIfNeeded("ggtern")
+installIfNeeded("hexbin")
+installIfNeeded("httr")
+installIfNeeded("igraph")
+installIfNeeded("jsonlite")
+installIfNeeded("magrittr")
+installIfNeeded("MASS")
+installIfNeeded("microbenchmark")
+installIfNeeded("phangorn")
+installIfNeeded("plotly")
+installIfNeeded("plotrix")
+installIfNeeded("profvis")
+installIfNeeded("robustbase")
+installIfNeeded("RColorBrewer")
+installIfNeeded("Rphylip")
+installIfNeeded("rvest")
+installIfNeeded("seqinr")
+installIfNeeded("stringi")
+installIfNeeded("taxize")
+installIfNeeded("testthat")
+installIfNeeded("xml2")
+
+# =    3  Bioconductor packages  ===============================================
+
+installIfNeeded("Biobase",       s = "BIO")
+installIfNeeded("biomaRt",       s = "BIO")
+installIfNeeded("Biostrings",    s = "BIO")
+installIfNeeded("DECIPHER",      s = "BIO")
+installIfNeeded("GEOquery",      s = "BIO")
+installIfNeeded("GOSim",         s = "BIO")
+installIfNeeded("limma",         s = "BIO")
+installIfNeeded("msa",           s = "BIO")
+installIfNeeded("org.Sc.sgd.db", s = "BIO")
+installIfNeeded("prada",         s = "BIO")
+installIfNeeded("topGO",         s = "BIO")
+
+
+# =    4  Other package sources  ===============================================
+
+# Using sources other than CRAN or Bioconductor to download general-purpose
+# programs that run on your computer is not generally recommended.
+
+
+# =    5  Updating packages  ===================================================
+
+# From time to time, update CRAN packages with the following command ...
+
+update.packages()
+
+# ... and also update Bioconductor packages as follows:
+
+BiocManager::install()
+
+# [END]