cell-specific markers

Description Usage Arguments Details Value Examples

This function permutes group labels among highest expressed M groups and then computes new OVESEG-test statistics.

1
2
3

OVEtstatPermTopM(y, group, groupOrder, M, weights = NULL,
  alpha = "moderated", NumPerm = 999, seed = 111,
  BPPARAM = bpparam())

`y`	a numeric matrix containing log-expression or logCPM (log2-counts per million) values. Data frame or SummarizedExperiment object will be internally coerced into a matrix. Rows correspond to probes and columns to samples. Missing values are not permitted.
`group`	categorical vector or factor giving group membership of columns of y. At least two categories need to be presented.
`groupOrder`	an integer matrix with each row giving group indexes ordered based on decreasing expression levels .
`M`	an integer indicating the number of groups being permutated. The range is [2,K], where K is the total number of groups.
`weights`	optional numeric matrix containing prior weights for each spot.
`alpha`	parameter specifying within-group variance estimator to be used. 'moderated': empirical Bayes moderated variance estimator as used in `eBayes`. Numeric value: a constant value added to pooled variance estimator (α + σ). NULL: no estimator; all variances are set to be 1.
`NumPerm`	an integer specifying the number of permutation resamplings (default 999).
`seed`	an integer seed for the random number generator.
`BPPARAM`	a BiocParallelParam object indicating whether parallelization should be used for permutation resamplings. The default is bpparam().

Top M expressed groups will be involved in permutation. There are C_K^M probe patterns in which probes are highly expressed in certain M groups among the total K groups. Probes of the same pattern share the same shuffled labels.

To improve the time efficiency, some functions within permutation loops are implemented using c++.

a list containing the following components:

`tstat.perm`	a numeric matrix with each column giving OVESEG-test statistics over the expressions after one permutation.
`topidx.perm`	a integer matrix with each colulmn giving the highest expressed group index over the expressions after one permutation.

data(RocheBT)
ppnull <- postProbNull(RocheBT$y, RocheBT$group, detail.return = TRUE)
rperm <- OVEtstatPermTopM(RocheBT$y, RocheBT$group, ppnull$groupOrder, M=2,
                         NumPerm=99, BPPARAM=BiocParallel::SerialParam())
## Not run: 
#parallel computing
rperm <- OVEtstatPermTopM(RocheBT$y, RocheBT$group, ppnull$groupOrder, M=2,
                         NumPerm=99, BPPARAM=BiocParallel::SnowParam())

## End(Not run)