chemodiv (version 0.3.1)

molNet: Generate a molecular network with some properties

Description

Function to generate a molecular network object, and some basic properties of the network.

Usage

molNet(compDisMat, npcTable = NULL, cutOff = "median")

Value

List with a (tbl_graph) graph object, the number of compounds, number of NPC pathways and a measure of the modularity of the network (see modularity).

Arguments

compDisMat

Compound dissimilarity matrix, as calculated by compDis. Note that the supplied dissimilarity matrix is transformed into a similarity matrix, and this is what cutOff values are set for. Note also that compDis always outputs a list of one or more matrices, while molNet requires a single matrix as input. Therefore, a specific matrix has to be selected from this list, as compDisOutput$matrix.

npcTable

A data frame generated by NPCTable can be supplied for calculations of the number of NPC pathways and network modularity.

cutOff

Cut-off value for compound similarities. Any similarity lower than this value will be set to zero when the network is generated, which strongly affects the look of the network. The value can be set manually to any value between 0 and 1; to the median similarity value from the compDisMat; or, if an NPCTable is supplied, to minPathway, the lowest within-pathway similarity (which allows all within-NPC-pathway similarities to be kept).

Details

Molecular networks can be used to illustrate the biosynthetic/structural similarity of phytochemical compounds in a sample, while simultaneously visualizing their relative concentrations. molNet creates the network, and molNetPlot can subsequently be used to create a plot of the network.

Examples

data(minimalCompDis)
data(minimalNPCTable)
molNet(minimalCompDis, minimalNPCTable, cutOff = 0)

data(alpinaCompDis)
molNet(compDisMat = alpinaCompDis, cutOff = 0.75)