weighted.score.multiple.vertex-methods {RANKS}

Multiple vertex score functions - weighted version

Description

Methods to compute weighted score functions for multiple vertices of the graph

Usage

## S4 method for signature 'graph'
NN.w.score(RW, x, x.pos, w, norm = TRUE)
## S4 method for signature 'matrix'
NN.w.score(RW, x, x.pos, w, norm = TRUE)
## S4 method for signature 'graph'
KNN.w.score(RW, x, x.pos, w, k = 3, norm = TRUE)
## S4 method for signature 'matrix'
KNN.w.score(RW, x, x.pos, w, k = 3, norm = TRUE)
## S4 method for signature 'graph'
eav.w.score(RW, x, x.pos, w, auto = FALSE, norm = TRUE)
## S4 method for signature 'matrix'
eav.w.score(RW, x, x.pos, w, auto = FALSE, norm = TRUE)

Arguments

RW

matrix. It must be a kernel matrix or a symmetric matrix expressing the similarity between nodes

x

vector of integer. Indices corresponding to the elements of the RW matrix for which the score must be computed

x.pos

vector of integer. Indices of the positive elements of the RW matrix

k

integer. Number of the k nearest neighbours to be considered

w

vector of numeric. Its elements represent the initial likelihood that the nodes of the graph belong to the class under study. The elements of w correspond to the columns of RW and the length of w and the number of columns of RW must be equal.

auto

boolean. If TRUE the components K(x,x) + K(x_i,x_i) are computed, otherwise are discarded (default)

norm

boolean. If TRUE (def.) the scores are normalized between 0 and 1.

Details

The methods compute the weighted scores for multiple vertices according to the weighted version of NN, KNN, and Empirical Average score. Note that the argument x indicates the set of nodes for which the score must be computed. The vector x represents the indices of the rows of the matrix RW corresponding to the vertices for which the scores must be computed. If x = 1:nrow(RW) the scores for all the vertices of the graph are computed.

Value

NN.w.score: a numeric vector with the weighted NN scores of the vertices. The names of the vector correspond to the indices x

KNN.score: a numeric vector with the weighted KNN scores of the vertices. The names of the vector correspond to the indices x

eav.score: a numeric vector with the weighted Empirical Average score of the vertices. The names of the vector correspond to the indices x

Methods

signature(RW = "graph")

NN.w.score computes the weighted NN score for multiple vertices using a graph of class graph (hence including objects of class graphAM and graphNEL from the package graph)

KNN.w.score computes the weighted KNN score for multiple vertices using a graph of class graph (hence including objects of class graphAM and graphNEL from the package graph)

eav.w.score computes the weighted Empirical Average score for multiple verticesusing a graph of class graph (hence including objects of class graphAM and graphNEL from the package graph)

signature(RW = "matrix")

NN.w.score computes the weighted NN score for multiple vertices using a kernel matrix or a symmetric matrix expressing the similarity between nodes

KNN.w.score computes the weighted KNN score for multiple vertices using a kernel matrix or a symmetric matrix expressing the similarity between nodes

eav.w.score computes the weighted Empirical Average score multiple for vertices using a kernel matrix or a symmetric matrix expressing the similarity between nodes

References

Giorgio Valentini, Giuliano Armano, Marco Frasca, Jianyi Lin, Marco Mesiti, and Matteo Re RANKS: a flexible tool for node label ranking and classification in biological networks Bioinformatics first published online June 2, 2016 doi:10.1093/bioinformatics/btw235

Re M, Mesiti M, Valentini G: A fast ranking algorithm for predicting gene functions in biomolecular networks. IEEE ACM Trans Comput Biol Bioinform 2012, 9(6):1812-1818.

See Also

Methods for scoring a single vertex - weighted version Methods for scoring multiple vertices

Examples

# Computation of scores using STRING data with respect to 
# the FunCat category 11.02.01 rRNA synthesis 
library(bionetdata);
data(Yeast.STRING.data);
data(Yeast.STRING.FunCat);
labels <- Yeast.STRING.FunCat[,"11.02.01"];
n <- length(labels);
ind.pos <- which(labels==1);
# NN-scores computed directly on the STRING matrix 
s <- NN.w.score(Yeast.STRING.data, 1:n, ind.pos, w=labels);
# Weighted NN-scores computed directly on the STRING matrix 
# using this time random weights for the value of positive nodes
w <- runif(n);
s <- NN.w.score(Yeast.STRING.data, 1:n, ind.pos, w=w);

# Weighted NN-scores computed on the 1 step and 2-step random walk kernel matrix
K <- rw.kernel(Yeast.STRING.data);
sK <- NN.w.score(K, 1:n, ind.pos, w);
K2 <- p.step.rw.kernel(K, p=2);
sK2 <- NN.w.score(K2, 1:n, ind.pos, w);