Test correspondance between predicted subgroups and known factors

Test correspondance between predicted subgroups and known factors

# S4 method for DownSamplingConsensusPartition
test_to_known_factors(object, k, known = get_anno(object),
    p_cutoff = 0.05, verbose = FALSE)

Arguments

object

A DownSamplingConsensusPartition-class object.

k

Number of subgroups. It uses all k if it is not specified.

known

A vector or a data frame with known factors. By default it is the annotation table set in consensus_partition_by_down_sampling.

p_cutoff

Cutoff for p-values for the class prediction. Samples with p-value higher than it are omit.

verbose

Whether to print messages.

Details

The test is performed by test_between_factors between the predicted classes and user's annotation table.

Value

A data frame with the following columns:

  • number of samples used to test after filtered by p_cutoff,

  • p-values from the tests,

  • number of subgroups.

Author

Zuguang Gu <z.gu@dkfz.de>

Examples

data(golub_cola_ds)
test_to_known_factors(golub_cola_ds, k = 3)
#>             n_sample ALL.AML(p-value) k
#> ATC:skmeans       70     1.873289e-09 3
test_to_known_factors(golub_cola_ds)
#>             n_sample ALL.AML(p-value) k
#> ATC:skmeans       71     2.248050e-04 2
#> ATC:skmeans       70     1.873289e-09 3
#> ATC:skmeans       67     1.550624e-09 4
#> ATC:skmeans       54     2.169022e-08 5
#> ATC:skmeans       49     3.783491e-06 6