Skip to contents

Compute local PWMs for a set of sequences given a PSSM matrix

compute_local_pwm.Rd

compute the local PWM for each position in every sequence. The edges of each sequences would become NA.

Usage

compute_local_pwm(
  sequences,
  pssm,
  spat = NULL,
  spat_min = 0,
  spat_max = NULL,
  bidirect = TRUE,
  prior = 0.01,
  return_list = FALSE
)

Arguments

sequences

a vector of sequences

pssm

a PSSM matrix or data frame. The columns of the matrix or data frame should be named with the nucleotides ('A', 'C', 'G' and 'T').

spat

a data frame with the spatial model (as returned from the $spat slot from the regression). Should contain a column called 'bin' and a column called 'spat_factor'.

spat_min

the minimum position to use from the sequences. The default is 1.

spat_max

the maximum position to use from the sequences. The default is the length of the sequences.

bidirect

is the motif bi-directional. If TRUE, the reverse-complement of the motif will be used as well.

prior

a prior probability for each nucleotide.

return_list

Logical. If TRUE, returns a list with one vector per sequence (useful for sequences of different lengths). If FALSE, returns a matrix (requires all sequences to have the same length). Default is FALSE.

Value

If return_list is FALSE: a matrix with length(sequences) rows and max(nchar(sequences)) columns with the local PWM for each sequence in each position. If return_list is TRUE: a list with length(sequences) elements, where each element is a numeric vector of local PWM scores for the corresponding sequence.

Examples

if (FALSE) { # \dontrun{
res <- regress_pwm(cluster_sequences_example, cluster_mat_example[, 1])

# Return matrix (sequences must have same length)
pwm <- compute_local_pwm(cluster_sequences_example, res$pssm, res$spat)
head(pwm)

# Return list (allows sequences of different lengths)
pwm_list <- compute_local_pwm(cluster_sequences_example, res$pssm, res$spat, return_list = TRUE)
pwm_list[[1]]

# Using a motif from MOTIF_DB
hnf1a_pwm <- compute_local_pwm(
    cluster_sequences_example,
    as.matrix(MOTIF_DB["JASPAR.HNF1A"]),
    return_list = TRUE
)
hnf1a_pwm[[1]]
} # }