Description Usage Arguments Value Author(s) References See Also Examples
Computation of the pseudo mrp-value from a resampling based feature
ranking strategy. qtl
represents the fraction of presumedly
informative features. The decision is based on the average rank across all
resampling steps. 1-qtl
represents the fraction of features that serves
to estimate the null distribution of ranks (i.e. ranks of uninformative variables).
1 | fs.mrpval(x,qtl=0.75)
|
x |
A list returned from |
qtl |
A numeric value of probability with values in [0,1]. |
A list with components:
stats |
Original feature ranking statistics. |
fs.rank |
Feature ranking vector. |
fs.order |
Feature order vector. |
sdrank |
Feature rank standard deviation. |
mrpval |
Individual feature mrp-value. |
Ug |
Uninformative variables. |
nnull |
Total number of uninformative variables. |
qtl |
Quantile |
David Enot dle@aber.ac.uk and Wanchang Lin wll@aber.ac.uk
Zhang, C., Lu,X. and Zhang, X. (2006). Significance of Gene Ranking for Classification of Microarray Samples. IEEE/ACM Transactions on Computational Biology and Bioinformatics, VOL. 3, NO. 3, pp. 312-320.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 | ## load abr1
data(abr1)
y <- factor(abr1$fact$class)
x <- preproc(abr1$pos , y=y, method=c("log10","TICnorm"),add=1)[,110:500]
## Select classes 1 and 2
dat <- dat.sel(x, y, choices=c("1","2"))
x <- dat$dat[[1]]
y <- dat$cl[[1]]
## partitioning
pars <- valipars(sampling="boot",niter=2,nreps=5)
tr.idx <- trainind(y,pars=pars)
## multiple rankings using AUC
z <- feat.rank.re(x,y,method="fs.auc",pars = pars,tr.idx=tr.idx)
## Compute stability mr-p value using the 25% worst features as irrelevant
res <- fs.mrpval(z,qtl=0.75)
## print content of res
names(res)
## list of features to form the null distribution of ranks
print(res$Ug)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.