plotGraph: Plotting igraph objects with information from a...

In microbiome/miaViz: Microbiome Analysis Plotting and Visualization

Plotting igraph objects with information from a SummarizedExperiment

Description

plotGraph plots an igraph object with additional information matched from a SummarizedExperiment object for the nodes only. Information on the edges have to provided manually.

Usage

plotColGraph(x, y, ...)

plotRowGraph(x, y, ...)

## S4 method for signature 'ANY,SummarizedExperiment'
plotColGraph(
  x,
  y,
  show.label = show_label,
  show_label = FALSE,
  add.legend = add_legend,
  add_legend = TRUE,
  layout = "kk",
  edge.type = edge_type,
  edge_type = c("fan", "link", "arc", "parallel"),
  edge.colour.by = edge_colour_by,
  edge_colour_by = NULL,
  edge.width.by = edge_width_by,
  edge_width_by = NULL,
  colour.by = colour_by,
  colour_by = NULL,
  shape.by = shape_by,
  shape_by = NULL,
  size.by = size_by,
  size_by = NULL,
  assay.type = by_exprs_values,
  by_exprs_values = "counts",
  other.fields = other_fields,
  other_fields = list(),
  ...
)

## S4 method for signature 'SummarizedExperiment,missing'
plotColGraph(x, y, name = "graph", ...)

## S4 method for signature 'ANY,SummarizedExperiment'
plotRowGraph(
  x,
  y,
  show.label = show_label,
  show_label = FALSE,
  add.legend = add_legend,
  add_legend = TRUE,
  layout = "kk",
  edge.type = edge_type,
  edge_type = c("fan", "link", "arc", "parallel"),
  edge.colour.by = edge_colour_by,
  edge_colour_by = NULL,
  edge.width.by = edge_width_by,
  edge_width_by = NULL,
  colour.by = colour_by,
  colour_by = NULL,
  shape.by = shape_by,
  shape_by = NULL,
  size.by = NULL,
  assay.type = by_exprs_values,
  by_exprs_values = "counts",
  other.fields = other_fields,
  other_fields = list(),
  ...
)

## S4 method for signature 'SummarizedExperiment,missing'
plotRowGraph(x, y, name = "graph", ...)

Arguments

x, y	a graph object and a SummarizedExperiment object or just a SummarizedExperiment. For the latter object a graph object must be stored in metadata(x)$name.
...	additional arguments for plotting. See mia-plot-args for more details i.e. call help("mia-plot-args")
show.label	Logical scalar, integer vector or character vector If a logical scalar is given, should tip labels be plotted or if a logical vector is provided, which labels should be shown? If an integer or character vector is provided, it will be converted to a logical vector. The integer values must be in the range of 1 and number of nodes, whereas the values of a character vector must match values of a label or name column in the node data. In case of a character vector only values corresponding to actual labels will be plotted and if no labels are provided no labels will be shown. (Default: FALSE)
show_label	Deprecated. Use show.label instead.
add.legend	Logical scalar. Should legends be plotted? (Default: TRUE)
add_legend	Deprecated. Use add.legend instead.
layout	Character scalar. Layout for the plotted graph. See ggraph for details. (Default: "kk")
edge.type	Character scalar. Type of edge plotted on the graph. See geom_edge_fan for details and other available geoms. (Default: "fan")
edge_type	Deprecated. Use edge.type instead.
edge.colour.by	Character scalar. Specification of an edge metadata field to use for setting colours of the edges. (Default: NULL)
edge_colour_by	Deprecated. Use edge.colour.by instead.
edge.width.by	Character scalar. Specification of an edge metadata field to use for setting width of the edges. (Default: NULL)
edge_width_by	Deprecated. Use edge.width.by instead.
colour.by	Character scalar. Specification of a column metadata field or a feature to colour graph nodes by, see the by argument in ?retrieveCellInfo for possible values. (Default: NULL)
colour_by	Deprecated. Use colour.by instead.
shape.by	Character scalar. Specification of a column metadata field or a feature to shape graph nodes by, see the by argument in ?retrieveCellInfo for possible values. (Default: NULL)
shape_by	Deprecated. Use shape.by instead.
size.by	Character scalar. Specification of a column metadata field or a feature to size graph nodes by, see the by argument in ?retrieveCellInfo for possible values. (Default: NULL)
size_by	Deprecated. Use size.by instead.
assay.type	Character scalar. or integer scalar. Specifies which assay to obtain expression values from, for use in point aesthetics - see the exprs_values argument in ?retrieveCellInfo. (Default: "counts")
by_exprs_values	Deprecated. Use assay.type instead.
other.fields	Additional fields to include in the node information without plotting them.
other_fields	Deprecated. Use other.fields instead.
name	Character scalar. If x is a SummarizedExperiment the key for subsetting the metadata(x) to a graph object. (Default: "graph")

Details

Internally tidygraph and ggraph are used. Therefore, all graph types which can be converted by tidygraph::as_tbl_graph can be used.

Value

a ggtree plot

Examples

# data setup
library(mia)
data(GlobalPatterns)
data(col_graph)
data(row_graph)
data(row_graph_order)
metadata(GlobalPatterns)$col_graph <- col_graph

genus <- agglomerateByRank(GlobalPatterns,"Genus",na.rm=TRUE)
metadata(genus)$row_graph <- row_graph
order <- agglomerateByRank(genus,"Order",na.rm=TRUE)
metadata(order)$row_graph <- row_graph_order

# plot a graph independently
plotColGraph(col_graph,
             genus,
             colour.by = "SampleType",
             edge.colour.by = "weight",
             edge.width.by = "weight",
             show.label = TRUE)

# plot the graph stored in the object
plotColGraph(genus,
             name = "col_graph",
             colour.by = "SampleType",
             edge.colour.by = "weight",
             edge.width.by = "weight")
             

# plot a graph independently
plotRowGraph(row_graph,
             genus,
             colour.by = "Kingdom",
             edge.colour.by = "weight",
             edge.width.by = "weight")

# plot the graph stored in the object
plotRowGraph(genus,
             name = "row_graph",
             colour.by = "Phylum",
             edge.colour.by = "weight",
             edge.width.by = "weight")

                           
# plot a graph independently
plotRowGraph(row_graph_order,
             order,
             colour.by = "Kingdom",
             edge.colour.by = "weight",
             edge.width.by = "weight")

# plot the graph stored in the object and include some labels
plotRowGraph(order,
             name = "row_graph",
             colour.by = "Phylum",
             edge.colour.by = "weight",
             edge.width.by = "weight", 
             show.label = c("Sulfolobales","Spirochaetales",
                            "Verrucomicrobiales"))
                            
# labels can also be included via selecting specific rownames of x/y
plotRowGraph(order,
             name = "row_graph",
             colour.by = "Phylum",
             edge.colour.by = "weight",
             edge.width.by = "weight", 
             show.label = c(1,10,50))
             
# labels can also be included via a logical vector, which has the same length
# as nodes are present
label_select <- rep(FALSE,nrow(order))
label_select[c(1,10,50)] <-  TRUE
plotRowGraph(order,
             name = "row_graph",
             colour.by = "Phylum",
             edge.colour.by = "weight",
             edge.width.by = "weight",
             show.label = label_select)

microbiome/miaViz documentation built on Nov. 4, 2024, 12:25 a.m.