Compute KL divergence matrix for PSSM datasets

Usage

pssm_dataset_diff(dataset1, dataset2 = NULL, prior = 0.01)

Arguments

dataset1: First PSSM dataset. A data frame with columns 'motif', 'A', 'C', 'G', 'T'
dataset2: Optional second PSSM dataset with the same structure as dataset1. If provided, computes divergences between motifs in dataset1 and dataset2. If NULL (default), computes divergences between all motifs in dataset1.
prior: A prior probability added to each nucleotide frequency (default: 0.01)

Value

If dataset2 is NULL, returns a symmetric square matrix of KL divergences between all motifs in dataset1. If dataset2 is provided, returns a matrix with rows corresponding to motifs in dataset1 and columns to motifs in dataset2.

Examples

if (FALSE) { # \dontrun{
# KL divergences within a single dataset
dm <- pssm_dataset_diff(JASPAR_motifs)

# KL divergences between two datasets
dm2 <- pssm_dataset_diff(JASPAR_motifs, HOMER_motifs)
} # }

# KL divergences within MOTIF_DB (subset)
gata_motifs <- as.data.frame(MOTIF_DB["GATA", pattern = TRUE])
dm_gata <- pssm_dataset_diff(gata_motifs)

# Compare GATA motifs with CDX motifs using KL divergence
cdx_motifs <- as.data.frame(MOTIF_DB["CDX", pattern = TRUE])
dm_cross <- pssm_dataset_diff(gata_motifs, cdx_motifs)