Package 'Director'

append2List

Description

Appends a data frame containing additional relationship information to an existing List having 6 columns: source, target, description, value, sourcefc, targetfc. Order matters! For example, add a map of transcripts to genes to a List of miRNAs and their target transcripts so that the final List connects miRNAs -> transcripts -> genes.

Usage

append2List(List, appendList, description = "description",
  sourcefc = "sourcefc", targetfc = "targetfc", value = "value",
  target = "target", source = "source", appendMatch = TRUE)

Arguments

List	Data frame containing the necessary columns above. e.g. Formatted with createList function.
appendList	Data frame or matrix to append to List.
description	Column name of appendList corresponding to the descriptions to append.
sourcefc	Column name of appendList corresponding to the sourcefcs to append.
targetfc	Column name of appendList correspondig to the targetfcs to append.
value	Column name of appendList correspondig to the relationship values to append.
target	Column name of appendList corresponding to the targets to append.
source	Column name of appendList corresponding to the sources to append.
appendMatch	Filter and remove 1) rows in List that contain targets without a corresponding source in appendList, and 2) rows in appendList that contain sources without a corresponding target in List.

Value

a combined List.

Examples

tempList <- createList(data.frame(source=c("A","B","C"),
    target=c("D","E","G"),
    description=c("consonant","vowel","consonant"),
    value=runif(3,-1,1),
    sourcefc=runif(3,-2,2),
    targetfc=runif(3,-2,2)))
tempAppendList <- data.frame(source="D",target="I",
    description="vowel",value=runif(1,-1,1),
    sourcefc=runif(1,-2,2), targetfc=runif(1,-2,2))
append2List(tempList,tempAppendList) # Will combine only 1 row from each list.
append2List(tempList,tempAppendList, appendMatch=FALSE) # Will combine all rows

---

createList

Description

Take a subset of the input data frame or matrix corresponding to the required Sankey values.

Usage

createList(inputList, inputFC = NULL, node = "genes", fc = "foldChange",
  source = "source", target = "target", description = "description",
  value = "value", sourcefc = "sourcefc", targetfc = "targetfc")

Arguments

inputList	Data frame or matrix containing the necessary parameters described below.
inputFC	Data frame or matrix containing node names (source and target) and corresponding quantitative values. If this input is defined, then input-specific parameters 'node' and 'fc' should be defined. Inputs 'source', 'target', 'description' and 'value' are still referenced from inputList.
node	Column name of inputFC containing names to display of source and target nodes. Paths defined in inputList identify which nodes are sources and which are targets.
fc	Column name of inputFC containing quantitative values representing the nodes.
source	Column name of inputList containing names to display of starting nodes. Paths are drawn from these points to their corresponding target nodes.
target	Column name of inputList containing names to display of destination nodes. Paths are drawn to these point from their corresponding source nodes.
description	Optional column name of inputList containing additional information about connection, e.g. the gene name of a transcript target node, or family name of related target genes.
value	Column name of inputList containing quantitative values representing the relationship between sources and targets.
sourcefc	Column name of inputList containing quantitative values representing the sources.
targetfc	Column name of inputList containing quantitative values representing the targets.

Value

a data.frame List

Examples

nodevals <- runif(5,-2,2)
tempList <- data.frame(source=c("A","B","C","D"),
    target=c("C","D","E","E"),
    addedInfo=c("c","d","vowel","vowel"),
    relationValue=runif(4,-1,1),
    sourceValue=nodevals[1:4],
    targetValue=nodevals[c(3,4,5,5)])
tempFC <- data.frame(genes=c("A","B","C","D","E"), foldChange=runif(5,-2,2))
# inputList only
createList(tempList, description="addedInfo", value="relationValue", 
sourcefc="sourceValue", targetfc="targetValue")
# inputList and inputFC
createList(tempList, tempFC, value="relationValue",sourcefc="sourceValue",
targetfc="targetValue")

---

drawSankey

Description

Create an HTML document that can be viewed and saved to file. Diagram properties can be modified in this function, makeSankey() and initSankey().

Usage

drawSankey(List, height = NULL, legendfont = "sans-serif",
  legendsize = 12, width = 1000, caption = "Sankey figure",
  nodeValue = "node values", pathValue = "path values", directory = NULL)

Arguments

List	Data frame containing 6 columns: source, target, description, value, sourcefc, targetfc.
height	Pixel height of the figure to draw. If empty, the figure will be given a pixel height proportional to the number of rows in List up to a maximum 1800px or minimum of 300px. These can be overridden by defining this parameter.
legendfont	Font of the legend text.
legendsize	Font size of the legend text.
width	Pixel width of the figure to draw. By default, 1000px.
caption	Sankey figure caption. HTML formatting is possible.
nodeValue	Description of node scale in legend.
pathValue	Description of path scale in legend.
directory	Absolute path to output directory. If null, the working directory obtained from getwd() will be used. This is required if D3 and sankey JS files were downloaded with initSankey().

Value

HTML document containing diagram.

Examples

Level1 <- createList(poorprog$Level1)
Level2 <- createList(poorprog$Level2)
tempList <- append2List(Level1,Level2)
initSankey()
tempList2 <- makeSankey(tempList, averagePath=TRUE)
sankey <- drawSankey(tempList2)
library(htmltools) # can also be launched with
html_print(sankey)

---

filterNumeric

Description

Filter a quantitave column in List for minimum, maximum, or absolute value.

Usage

filterNumeric(List, column, min = NULL, max = NULL, absolute = NULL)

Arguments

List	Data frame containing 6 columns: source, target, description, value, sourcefc, targetfc.
column	Name of column in List to filter.
min	Minimum value to filter for in column.
max	Maximum value to filter for in column.
absolute	Absolute value to filter for in column.

Value

a filtered List

Examples

tempList <- createList(data.frame(source=c("A","B","C"),
    target=c("D","E","G"),
    description=c("consonant","vowel","consonant"),
    value=runif(3,-1,1),
    sourcefc=runif(3,-2,2),
    targetfc=runif(3,-2,2)))
filterNumeric(tempList,"sourcefc", absolute=0.5)
filterNumeric(tempList, "targetfc", max=0) # only take down-regulated targets

---

filterRelation

Description

Filter source-target relationships in List for a specific type: inversely related sourcefc-targetfc pairs only (inverseFC), positively related sourcefc-targetfc pairs only (correlatedFC), negative value scores only (inverseValue), or positive value scores only (correlatedValue). Default is to not apply any filtering.

Usage

filterRelation(List, relation = c("none", "inverseFC", "correlatedFC",
  "inverseValue", "correlatedValue"), sourcefc = "sourcefc",
  targetfc = "targetfc", value = "value")

Arguments

List	Data frame containing 6 columns: source, target, description, value, sourcefc, targetfc.
relation	One of: none, inverseFC, correlatedFC, inverseValue, correlatedValue. Default is none.
sourcefc	Column name of List corresponding to sourcefcs to filter.
targetfc	Column name of List correspondig to targetfcs to filter.
value	Column name of List correspondig to value to filter.

Value

a filtered List.

Examples

tempList <- createList(data.frame(source=c("A","B","C"),
    target=c("D","E","G"),
    description=c("consonant","vowel","consonant"),
    value=runif(3,-1,1),
    sourcefc=runif(3,-2,2),
    targetfc=runif(3,-2,2)))
filterRelation(tempList,"inverseValue")
filterRelation(tempList,"correlatedValue")
filterRelation(tempList,"inverseFC")
filterRelation(tempList,"correlatedFC")

---

filterSubset

Description

Filter up to two qualitative columns (source and target) in List for a subset of names.

Usage

filterSubset(List, sourceSubset = NULL, targetSubset = NULL,
  invert = FALSE, source = "source", target = "target",
  join = c("union", "intersect"))

Arguments

List	Data frame containing 6 columns: source, target, description, value, sourcefc, targetfc.
sourceSubset	Vector of source names to keep.
targetSubset	Vector of target names to keep.
invert	Take the inverse selection of defined subset.
source	Column name of List containing source names.
target	Column name of List containing target names.
join	If both subsets are defined, take either union or intersect of subsets found.

Value

a filtered List

Examples

tempList <- createList(data.frame(source=c("A","B","C"),
    target=c("D","E","G"),
    description=c("consonant","vowel","consonant"),
    value=runif(3,-1,1),
    sourcefc=runif(3,-2,2),
    targetfc=runif(3,-2,2)))
filterSubset(tempList,source="source", target="description", 
sourceSubset="C", targetSubset="consonant")
filterSubset(tempList,target="description", targetSubset="consonant")
filterSubset(tempList,target="description", targetSubset="consonant", invert=TRUE)

---

initSankey

Description

Internally generates supporting JavaScript and CSS files.

Usage

initSankey(pathOpacity = 0.2, pathHover = 0.5,
  font = "lato, helvetica, sans-serif", fontsize = NULL,
  fontsizeProportion = TRUE, d3js = NULL, sankeyjsFile = NULL,
  d3jsMethod = "auto", sankeyjsMethod = "auto")

Arguments

pathOpacity	Opacity of connecting path between nodes in the figure.
pathHover	Opacity of connecting path between nodes upon mouseover.
font	Font used for the node names and additional mouseover text in figure.
fontsize	Pixel font size used for the visible node names. Use to adjust range of font sizes (with proportions) or to set a single font size when fontsizeProportion is disabled.
fontsizeProportion	Boolean to enable/disable text being proportional to node widths. When enabled, all node names will appear with parameter fontsize.
d3js	Path to download latest zip version of D3 library. e.g. https://github.com/mbostock/d3/releases/download/v3.5.16/d3.zip. See http://www.d3js.org for more details. If NULL, will use version 3.5.16 currently installed with Director.
sankeyjsFile	Path to download sankey javascript file. If NULL, will use version installed with Director (https://raw.githubusercontent.com/d3/d3-plugins/master/sankey/sankey.js)
d3jsMethod	Function method to use to download D3 library. ?download.file for more detail on parameter.
sankeyjsMethod	Function method to use to download sankey script. ?download.file for more detail on parameter.

Value

global JavaScript and CSS files.

Examples

initSankey() # Generates supporting JavaScript and CSS files.

---

makeSankey

Description

Takes a list of source-target pairs and assigns colours to nodes and connections based on value, sourcefc and targetfc. Output is a list with List$reference = input List with additional description values, $valDomain = path values, $valRange = path colours, $targetDomain = target names, $targetRange = target node colours, $sourceDomain = source names, $sourceRange = source node colours.

Usage

makeSankey(List, averagePath = FALSE, nodeMin = "blue", nodeMax = "red",
  pathMin = "blue", pathMax = "red", noughtColor = "#f5f5f0",
  nought = 0, noughtPath = NULL, noughtPathColor = NULL)

Arguments

List	Data frame containing 6 columns: source, target, description, value, sourcefc, targetfc.
averagePath	Boolean to either keep List$value as-is, or calculate List$value for intermediary nodes (i.e. source nodes that were previously target nodes) as an average of previous path List$values.
nodeMin	Colour assigned to minimum node value.
nodeMax	Colour assigned to maximum node value.
pathMin	Colour assigned to minimum path value.
pathMax	Colour assigned to maximum path value.
noughtColor	Colour assigned to nought value.
nought	'Zero' value dividing node and paths into two distinct sets. i.e. positive and negative.
noughtPath	Optional parameter that sets a different 'zero' value for paths than for nodes.
noughtPathColor	Optional parameter that assigns a different colour to the path 'zero' value from the node 'zero' value.

Value

a list of data.frames and colour vectors.

Examples

tempList <- createList(data.frame(source=c("A","B","C"),
    target=c("D","E","G"),
    description=c("consonant","vowel","consonant"),
    value=runif(3,-1,1),
    sourcefc=runif(3,-2,2),
    targetfc=runif(3,-2,2)))
initSankey()
tempList2 <- makeSankey(tempList)

---

Analysis results of Yang et al.'s (2013) master microRNA regulatory network

Description

Data frames listing a set of genes differentially expressed between mesenchymal and three other serous ovarian cancer subtypes, eight key miRNAs predicted to target them, and significantly enriched pathways (FDR < 0.1). Each row contains a miRNA-gene/gene-pathway pair, a description, expression correlation (path values), and expression fold-change (node values).

Usage

ovca

Format

a list instance containing 2 data frames.

Value

data frame

Source

The Cancer Genome Atlas. Yang et al., 2013.

---

Analysis results for poor prognosis serous ovarian cancer

Description

Data frames listing a set of genes differentially expressed between long surviving (good prognosis) and short surviving (poor prognosis) cases, their putative targeting miRNAs, and significantly enriched pathways (FDR < 0.1). Each row contains a miRNA-gene/gene-pathway pair, a description, expression correlation (path values), and expression fold-change (node values).

Usage

ovca

Format

a list instance containing 2 data frames.

Value

data frame

Source

The Cancer Genome Atlas

---

writeSankey

Description

Save sankey figure as a simple HTML file accessible outside of R and shiny. Functions initSankey, makeSankey and drawSankey must be performed before this step to ensure a proper figure is saved.

Usage

writeSankey(name = NULL, title = NULL, directory = NULL)

Arguments

name	Name to give file. If no path given, the working directory OR path set in Director will be used. Same name will be given as the title.
title	Title of the HTML file produced. The file name is used by default.
directory	Absolute path to output directory. If null, the working directory obtained from getwd() will be used. If no absolute path is given (i.e. no "/" is grepped), it will assume a new folder will be created in the working directory.

Value

a dynamic HTML file in the specified directory that is readable in any internet browser so long as the 'www' subfolder is included.

Examples

Level1 <- createList(poorprog$Level1)
Level2 <- createList(poorprog$Level2)
tempList <- append2List(Level1,Level2)
initSankey() # initializes working directory
tempList2 <- makeSankey(tempList, averagePath=TRUE) # Calculate node and path values
sankey <- drawSankey(tempList2)
writeSankey("temp") # Save figure as the file 'temp.html' in working directory.

---