ggraph: Create a ggraph plot

In ggraph: An Implementation of Grammar of Graphics for Graphs and Networks

Create a ggraph plot

Description

This function is the equivalent of ggplot2::ggplot() in ggplot2. It takes care of setting up the plot object along with creating the layout for the plot based on the graph and the specification passed in. Alternatively a layout can be prepared in advance using create_layout and passed as the data argument. See Details for a description of all available layouts.

Usage

ggraph(graph, layout = "auto", ...)

create_layout(graph, layout, circular, ...)

## Default S3 method:
create_layout(graph, layout, ...)

## S3 method for class 'layout_ggraph'
create_layout(graph, ...)

## S3 method for class 'tbl_graph'
create_layout(graph, layout, circular = FALSE, ...)

Arguments

graph	The object containing the graph. See Details for a list of supported classes. Or a layout_ggraph object as returned from create_layout in which case all subsequent arguments is ignored.
layout	The type of layout to create. Either a valid string, a function, a matrix, or a data.frame (see Details)
...	Arguments passed on to the layout function.
circular	Should the layout be transformed into a radial representation. Only possible for some layouts. Defaults to FALSE

Details

Following is a short description of the different layout types available in ggraph. Each layout is further described in its own help pages. Any type of regular graph/network data can be represented as a tbl_graph object. Because of this the different layouts that can be applied to tbl_graph objects are quite diverse, but not all layouts makes sense to all types of graphs. It is up to the user to understand their data and choose an appropriate layout. For standard node-edge diagrams igraph defines a long range of different layout functions that are all available through the igraph layout where the specific layout is specified using the algorithm argument. In order to minimize typing all igraph algorithms can also be passed directly into the layout argument.

Any object that has an appropriate as_tbl_graph method can be passed into ggraph() and will automatically be converted underneath.

auto

The default layout. See layout_tbl_graph_auto() for further details

igraph

Use one of the internal igraph layout algorithms. The algorithm is specified using the algorithm argument. All strings accepted by the algorithm argument can also be supplied directly into layout. See layout_tbl_graph_igraph() for further details

dendrogram

Lays out the nodes in a tree-like graph as a dendrogram with leaves set at 0 and parents 1 unit above its tallest child. See layout_tbl_graph_dendrogram() for further details

manual

Lets the user manually specify the location of each node. See layout_tbl_graph_manual() for further details

linear

Arranges the nodes linearly or circularly in order to make an arc diagram. See layout_tbl_graph_linear() for further details

matrix

Arranges nodes on a diagonal thus preparing it for use with geom_edge_point() to make a matrix plot. See layout_tbl_graph_matrix() for further details

treemap

Creates a treemap from the graph, that is, a space-filing subdivision of rectangles showing a weighted hierarchy. See layout_tbl_graph_treemap() for further details

circlepack

Creates a layout showing a hierarchy as circles within circles. Conceptually equal to treemaps. See layout_tbl_graph_circlepack() for further details

partition

Create icicle or sunburst charts, where each layer subdivides the division given by the preceding layer. See layout_tbl_graph_partition() for further details

hive

Positions nodes on axes spreading out from the center based on node attributes. See layout_tbl_graph_hive() for further details

cactustree

Positions nodes as circles on the periphery of their parent circle. See layout_tbl_graph_cactustree() for further details

backbone

Layout optimised for highly connected small-world graphs such as social networks. See layout_tbl_graph_backbone() for further details

centrality

Place nodes around origin based on their centrality. See layout_tbl_graph_centrality() for further details

eigen

Spectral layout based on the eigenvector of a matrix representation of the graph. See layout_tbl_graph_eigen() for further details

fabric

Draw nodes as horizontal lines and connect them with vertical lines if an edge exists between them. See layout_tbl_graph_fabric() for further details

focus

Place nodes around a focus node based on their distance to that node. See layout_tbl_graph_focus() for further details

pmds

Layout based on multidimensional scaling of a set of pivot nodes, allowing MDS layout to be used on larger graphs. See layout_tbl_graph_pmds() for further details

stress

Layout based on stress minimisation with better stability than Kamada-Kawai layout. See layout_tbl_graph_stress() for further details

unrooted

Draws unrooted trees based on equal angle with optional equal daylight modification. See layout_tbl_graph_unrooted() for further details

htree

Draws binary trees as a space filling fractal. See layout_tbl_graph_htree() for further details

Alternatively a matrix or a data.frame can be provided to the layout argument. In the former case the first column will be used as x coordinates and the second column will by used as y coordinates, further columns are dropped. In the latter case the data.frame is used as the layout table and must thus contain a numeric x and y column.

Lastly a function can be provided to the layout argument. It will be called with the graph object as its first argument and any additional argument passed into ggraph()/create_layout(). The function must return either a data.frame or an object coercible to one and have an x and y column, or an object coercible to a tbl_graph. In the latter case the node data is extracted and used as layout (and must thus contain an x and y column) and the graph will be added as the graph attribute.

Value

For ggraph() an object of class gg onto which layers, scales, etc. can be added. For create_layout() an object inheriting from layout_ggraph. layout_ggraph itself inherits from data.frame and can be considered as such. The data.frame contains the node positions in the x and y column along with additional columns generated by the specific layout, as well as node parameters inherited from the graph. Additional information is stored as attributes to the data.frame. The original graph object is stored in the graph attribute and the circular attribute contains a logical indicating whether the layout has been transformed to a circular representation.

See Also

get_edges() for extracting edge information from the layout and get_con() for extracting path information.

Examples

require(tidygraph)
gr <- create_notable('bull')
layout <- create_layout(gr, layout = 'igraph', algorithm = 'kk')

ggraph documentation built on May 29, 2024, 2:32 a.m.