This function calculates the IVI of the desired nodes from a graph. #' A shiny app has also been developed for the calculation of IVI as well as IVI-based network visualization, which is accessible using the influential::runShinyApp("IVI") command. You can also access the shiny app online at https://influential.erc.monash.edu/.

ivi(
graph,
vertices = V(graph),
weights = NULL,
directed = FALSE,
mode = "all",
loops = TRUE,
d = 3,
scaled = TRUE
)

## Arguments

graph

A graph (network) of the igraph class.

vertices

A vector of desired vertices, which could be obtained by the V function.

weights

Optional positive weight vector for calculating weighted betweenness centrality of nodes as a requirement for calculation of IVI. If the graph has a weight edge attribute, then this is used by default. Weights are used to calculate weighted shortest paths, so they are interpreted as distances.

directed

Logical scalar, whether to directed graph is analyzed. This argument is ignored for undirected graphs.

mode

The mode of IVI depending on the directedness of the graph. If the graph is undirected, the mode "all" should be specified. Otherwise, for the calculation of IVI based on incoming connections select "in" and for the outgoing connections select "out". Also, if all of the connections are desired, specify the "all" mode. Default mode is set to "all".

loops

Logical; whether the loop edges are also counted.

d

The distance, expressed in number of steps from a given node (default=3). Distance must be > 0. According to Morone & Makse (https://doi.org/10.1038/nature14604), optimal results can be reached at d=3,4, but this depends on the size/"radius" of the network. NOTE: the distance d is not inclusive. This means that nodes at a distance of 3 from our node-of-interest do not include nodes at distances 1 and 2. Only 3.

scaled

Logical; whether the end result should be 1-100 range normalized or not (default is TRUE).

## Value

A numeric vector with the IVI values based on the provided centrality measures.

cent_network.vis

Other integrative ranking functions: comp_manipulate(), exir(), hubness.score(), ivi.from.indices(), spreading.score()

## Examples

if (FALSE) {
MyData <- coexpression.data
My_graph <- graph_from_data_frame(MyData)
GraphVertices <- V(My_graph)
My.vertices.IVI <- ivi(graph = My_graph, vertices = GraphVertices,
weights = NULL, directed = FALSE, mode = "all",
loops = TRUE, d = 3, scaled = TRUE)
}