combine: Combine structural and statistical 'AdjacencyMatrix' objects

In tnaake/MetNet: Inferring metabolic networks from untargeted high-resolution mass spectrometry data

Combine structural and statistical 'AdjacencyMatrix' objects

Description

The function 'combine' takes as input the structural and statistical 'AdjacencyMatrix' objects, created in former steps. It will access the assays 'binary' and 'consensus', adds them together and will report a connection between metabolites if the edge is present in both matrices.

'combine' returns an 'AdjacencyMatrix' containing this consensus matrix supported by the structural and statistical adjacency matrices (assay 'combine_binary'). The 'AdjacencyMatrix' object furthermore contains the assays from the statistical 'AdjacencyMatrix' and the combined assays from the structural 'AdjacencyMatrix', e.g. if the structural 'AdjacencyMatrix' has the assays 'group' and 'mass', the combine 'AdjacencyMatrix' object will contain the assays 'combine_group' and 'combine_mass' that have support from the structural and statistical 'AdjacencyMatrix' object.

Usage

combine(am_structural, am_statistical)

Arguments

am_structural	'AdjacencyMatrix' containing the 'numeric' structural adjacency matrix (assay 'binary') and other 'character' or 'numeric' structural and spectral similarity adjacency matrices (e.g. 'group', 'mass' or spectral similarity as 'ndotprodcut').
am_statistical	'AdjacencyMatrix' containing the assay 'consensus' and other 'numeric' adjacency matrices depending on the chosen statistical models

Details

The matrices from the assays 'binary' and 'consensus' will be added and an unweighted connection will be reported when the edges are respectively present in both 'binary' and 'consensus'.

Value

'AdjacencyMatrix' object containing the assays 'combine_binary' ('numeric' adjacency matrix), and the combined matrices derived from the structural 'AdjacencyMatrix' ('character' adjacency matrices).

The 'AdjacencyMatrix' object will also contain all other assays contained in 'am_structural' and 'am_statistical'.

Author(s)

Thomas Naake, [e-mail]

Examples

data("x_test", package = "MetNet")
x_test <- as.matrix(x_test)
transformation <- rbind(
    c("Monosaccharide (-H2O)", "C6H10O5", "162.0528234315"),
    c("Disaccharide (-H2O)", "C12H20O11", "340.1005614851"),
    c("Trisaccharide (-H2O)", "C18H30O15", "486.1584702945"))
transformation <- data.frame(group = transformation[, 1],
    formula = transformation[, 2],
    mass = as.numeric(transformation[, 3]))
     
## create AdjacencyMatrix object of type structural
am_struct <- structural(x_test, transformation, var = c("group", "mass"), 
    ppm = 10)

## create AdjacencyMatrix object of type statistical
x_test_cut <- as.matrix(x_test[, -c(1:2)])
am_stat <- statistical(x_test_cut, model = c("pearson", "spearman"),
    adjust = "bonferroni")
am_stat <- threshold(am_stat, type = "top2", args = list(n = 10))

## combine
combine(am_structural = am_struct, am_statistical = am_stat)

tnaake/MetNet documentation built on Nov. 4, 2024, 4:45 a.m.