build.wildcard.dictionary: Build dictionary for wildcard kernel

In ManuSetty/ChIPKernels: String kernels for ChIP-Seq

Description

This build a matrix of distances between all kmers and unique kmers, considering reverse complements and matches to degenerate nucleotide alphabet.

Usage

  build.wildcard.dictionary(kmer.len, mismatches, out.dir,
    alphabet = c("A", "C", "G", "T", "N", "S", "M", "R", "Y", "K", "W"),
    consecutive.mismatches = TRUE, no.cores = 1,
    verbose = TRUE)

Arguments

kmer.len	Length of kmers
mismatches	Number of mismatches
out.dir	Directory to write the diciontary in
alphabet	Nucleotide alphabet. By defualt using all single/double nucleotide combinations and N Note that the first four alphabets have to be c('A', 'C', 'G', 'T')
consecutive.mismatches	Logical indicating whether only consecutive mismatches should be allowed
no.cores	Number of cores of parallel processing
verbose	Logical to show logs

Value

Creates a file containing the following elements: pairwise.kmers Distance matrix and kmer.mapping Mapping of reverse complements and kmers to features The filename is constructed based on kmer length and kernel used and has the structure: positional_wildcard_dict_kmer<kmer.len_mismatchesmismatches_alphalength (alphabet).Rdata

ManuSetty/ChIPKernels documentation built on May 7, 2019, 2:46 p.m.