Below are the basics of the package.

  1. All graphs must be initialised with the sigmajs() function.
  2. All the functions of the package start with sg_.
  3. Functions starting in sg_get_ are helpers that do not take a sigmajs object as input.
  4. All proxies end in _p.
  5. Functions are pipe-friendly (%>%).


The variable names must follow sigma.js convention.

Nodes must ideally include, at least:

  • id unique id of node, note that this does not have to be numerical
  • size size of node
  • color color of node

These are passed to sg_nodes. Note that sigma.js requires passing x and y, if omitted the function will randomly assign x and y positions.

Edges must include, at least:

  • id unique id of edge
  • source source of edge
  • target target of edge

source and target refer to the nodes id

Remember to always follow the naming convention.

Convenience functions

The package comes with two convenience functions to generate data, mainly to be used for testing.

  • sg_make_nodes to generate nodes, you can specify the number of nodes you want.
  • sg_make_edges to generate edges based on the nodes, you can specify the number of edges you want.


Above I state that you must ideally pass size and color to the nodes. I say “ideally” because, you actually can omit color if you then pass default color using sg_settings.

There a lot of settings to pass, see the official wiki for a list of all settings.

Your first graphs

# generate data using convenience functions
nodes <- sg_make_nodes()
edges <- sg_make_edges(nodes)

sigmajs() %>%
  sg_nodes(nodes, id, size, color) %>%
  sg_edges(edges, id, source, target)

Or as stated in the Settings section, you can omit size and color but you then must pass defaults using sg_settings.

sigmajs() %>%
  sg_nodes(nodes, id, size) %>%
  sg_edges(edges, id, source, target) %>%
  sg_settings(defaultNodeColor = "#c9423f") # pass default color

Browse the documentation to see how to give your graphs neat layouts, animate them, see how the package works with shiny and more.