do.RWR {RANKS}

Random walk with restart cross-validation experiments with multiple classes

Description

High level function to perform random walk with restart cross-validation experiments with multiple classes.

Usage

do.RWR(gamma = 0.6, tmax = 1000, eps = 1e-10, k = 5, stratified=TRUE, filter = TRUE, 
 seed = 1, data.dir, labels.dir, output.dir, data, labels)

Arguments

gamma

restart parameter (def: 0.6)

tmax

maximum number of iterations (def: 1000)

eps

maximum allowed difference between the computed probabilities at the steady state (def. 1e-10)

k

number of folds for the cross validation (def. 5)

stratified

boolean. If TRUE (def.) stratified CV is performed otherwise vanilla CV is done

filter

if TRUE (def) the adjacency matrix is sparsified otherwise not

seed

seed of the random generator for the generation of the folds (def: 1)

data.dir

relative path to directory where the adjiacency matrix is stored

labels.dir

relative path to directory where the label matrix is stored

output.dir

relative path to directory where the results are stored

data

name of the data set to loaded (without rda extension). It must be an .rda file containing the adjiacency matrix of the graph. It assumes that it is in the data.dir directory

labels

name of the target labels (without rda extension). It must be an .rda file containing the label matrix of the examples. It assumes that it is in the labels.dir directory

Details

High level function to perform cross-validation experiments with multiple classes using RWR.

It performs a k fold CV on a given data set, and output scores, AUC and Precision at a given recall results for multiple classes.

Graph data are read from a matrix representing the adjiacency matrix of the graph stored as a .rda file. The labels are read from a matrix having examples as rows and classes as columns stored as a .rda file. If M is the label matrix, then M[i,j]=1, if example i is annotated with class j, otherwise M[i,j] = 0.

Results are included in matrices representing Scores, AUC and precision at a given recall results stored as .rda files.

Value

3 rda files stored in the output.dir directory:

Scores results

A matrix with examples on rows and classes on columns representing the computed scores for each example and for each considered class

AUC results

AUC results files computed through AUC.single.over.classes from the package PerfMeas

Precision at given recall results

computed through precision.at.multiple.recall.level.over.classes from the package PerfMeas.

The name of the Score file starts with Score, of the AUC file with AUC, and of the Precision at given recall file with PXR. Other learning parameters are appended to the name of the file. All the results .rda files are stored in the Results directory (that must exist in advance).

See Also

RWR, multiple.RW.cv, do.RW

Examples

# Yeast prediction of 177 FunCat classes by 5-fold cross validation 
# using 3 steps of Random walk with restart and STRING data. 
# data obtained from the bionetdata package from CRAN
# See the AUC and Precision/recall results in the Results directory
library(bionetdata);
dd=tempdir();
rr=tempdir();
data(Yeast.STRING.data);
data(Yeast.STRING.FunCat);
save(Yeast.STRING.data, file=paste(dd,"/net.rda", sep=""));
save(Yeast.STRING.FunCat, file=paste(dd,"/labels.rda", sep=""));
do.RWR(tmax = 3, k = 5, filter = FALSE, seed = 1, data.dir=dd, labels.dir=dd, 
output.dir=rr, data="/net", labels="/labels");
# the same experiment, but the iterations are repeated till to convergence 
# (this can require a quite long time ...)
do.RWR(tmax = 1000, k = 5, eps = 1e-5, filter = FALSE, seed = 1, data.dir=dd, 
labels.dir=dd, output.dir=rr, data="/net", labels="/labels");