Entropy-Based Feature Selection Algorithms — filter

This function is a wrapper over Entropy-Based Feature Selection Algorithms for k-mer data.

filter_ig(target, kmers, method)

Arguments

target: a numeric response variable
kmers: a matrix of kmers with named columns or an object obtained via generate_kmer_data function.
method: a character name of a filter type. One of "infogain", "gainratio" or "symuncert". For more details see information_gain.

Value

a numeric vector of named p-values corresponding to k-mers in the feature space.

Details

This function uses information_gain

Examples

n_seq <- 20
sequence_length <- 20
alph <- letters[1:4]
motifs <- generate_motifs(alph, 4, 4, 4, 6)
kmers <- generate_kmer_data(n_seq, sequence_length, alph,
                            motifs, n_injections = 4)
target <- get_target_additive(kmers)
filter_ig(target, kmers, "infogain", 0.01)
#> Error in filter_ig(target, kmers, "infogain", 0.01): unused argument (0.01)