Description Usage Arguments Details Value Author(s) See Also Examples

This routine performs further computations on a list of `feat.rank.re`

objects contained in a
`mfr.obj`

object (see `ftrank.agg`

). If only one `feat.rank.re`

result from `feat.rank.re`

is analyzed, it is easier to pass it first to `ftrank.agg`

(see example).Two calculations are made: 1) Computation of the pseudo mrp-value from the resampling based feature (see `fs.mrpval`

) and 2) adjusted p values if p values have calculated (see `p.adjust`

).

1 | ```
summ.ftrank(lclas, lmod = NULL, qtl = 0.25, padjust = "fdr")
``` |

`lclas` |
mfr.obj object - See details in |

`lmod` |
List of models to be considered in lclas |

`qtl` |
Quantile - See details in |

`padjust` |
p value adjustement method - See details in |

The resulting list as two component: one is equal to the total number of `feat.rank.re`

objects (i.e one resampling experiment) and one field is a table that summarises each `feat.rank.re`

(as for `ftrank.agg`

). In the first component, each table may have a different number of columns depending if the FR method outputs p-values or not:

- Stat:
Original statistics calculated on the overall data.

- Rank:
Original feature rank calculated on the overall data.

- pval:
Original feature p-value calculated on the overall data (optional).

- pval-xxx:
Feature adjusted p-value by method xxx if p-value available (i.e. previous column).

- mrpval-xxx:
Pseudo multiple resampling p-value using a given

`qtl`

value xxx.- AvgRk:
Average feature rank calculated from the ranks found for each training data partition.

- SdevRk:
Associated feature rank standard deviation calculated from the ranks found for each training data partition.

`mfr.sum`

object:

`frsum` |
List of tables corresponding to each |

`frdef` |
Summary of each |

David Enot [email protected]

`fs.mrpval`

, `tidy.ftrank`

, `p.adjust`

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | ```
data(abr1)
y <- factor(abr1$fact$class)
x <- preproc(abr1$pos , y=y, method=c("log10","TICnorm"),add=1)[,110:500]
## Select classes 1 and 2
dat <- dat.sel1(x, y, pwise="1",mclass=NULL,pars=valipars(sampling="boot",niter=2,nreps=5))
reswelch = feat.rank.re(dat[[1]],method="fs.welch")
mfr=ftrank.agg(reswelch)
print(mfr$frdef)
frsum=summ.ftrank(mfr,lmod=1,qtl=.3)
## print the FR components
print(frsum$frdef)
## have a look at the first 5 variables
print(frsum$frsum[[1]][1:5,])
``` |

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.