Hive plots were invented by Martin Krzywinski as a perceptually uniform and scalable alternative to standard node-edge layouts. In hive plots nodes are positioned on axes radiating out from a center based on their own information e.g. membership of a class, size of neighborhood, etc. Edges are then drawn between nodes as bezier curves. As the placement of nodes is not governed by convoluted algorithms but directly reflects the qualities of the nodes itself the resulting plot can be easier to interpret as well as compare to other graphs.
layout_tbl_graph_hive( graph, axis, axis.pos = NULL, sort.by = NULL, divide.by = NULL, divide.order = NULL, normalize = TRUE, center.size = 0.1, divide.size = 0.05, use.numeric = FALSE, offset = pi/2, split.axes = "none", split.angle = pi/6, circular = FALSE )
The node attribute to use for assigning nodes to axes
The relative distance to the prior axis. Default
The node attribute to use for placing nodes along their axis.
An optional node attribute to subdivide each axis by.
The order the axis subdivisions should appear in
Logical. Should axis lengths be equal or reflect the number
of nodes in each axis. Defaults to
The size of the blank center, that is, the start position of the axes.
The distance between subdivided axis segments.
Logical, If the
Change the overall rotation of the hive plot by changing the offset of the first axis.
Should axes be split to show edges between nodes on the same axis? One of:
The angular distance between the two axes resulting from a split.
A data.frame with the columns
circular as well as any information stored as node variables in the
In order to be able to draw all edges without edges crossing axes you should not assign nodes to axes based on a variable with more than three levels.
Krzywinski, M., Birol, I., Jones, SJM., and Marra, MA. (2012). Hive plots-rational approach to visualizing networks. Brief Bioinform 13 (5): 627-644. https://doi.org/10.1093/bib/bbr069