Function to generate a k-mer periodicity track

This function takes a set of GRanges in a genome, recover the corresponding sequences and divides them using a sliding window. For each sub-sequence, it then computes the PSD value of a k-mer of interest at a chosen period, and generates a linear .bigWig track from these values.

getPeriodicityTrack(
  genome = NULL,
  granges,
  motif = "WW",
  period = 10,
  BPPARAM = setUpBPPARAM(1),
  extension = 1000,
  window_size = 100,
  step_size = 2,
  range_spectrum = seq(5, 50),
  smooth_track = 20,
  bw_file = NULL
)

Arguments

genome	DNAStringSet, BSgenome or genome ID
granges	GRanges object
motif	character, k-mer of interest.
period	Integer, the period of the k-mer to study (default=10).
BPPARAM	split the workload over several processors using BiocParallel
extension	Integer, the width the GRanges are going to be extended to (default 1000).
window_size	Integer, the width of the bins to split the GRanges objects in (default 100).
step_size	Integer, the increment between bins over GRanges (default 2).
range_spectrum	Numeric vector, the distances between nucleotides to take into consideration when performing Fast Fourier Transform (default seq_len(50)).
smooth_track	Integer, smooth the resulting track
bw_file	character, the name of the output bigWig track

Value

Rlelist and a bigWig track in the working directory.

Examples

data(ce11_proms)
track <- getPeriodicityTrack(
    genome = 'BSgenome.Celegans.UCSC.ce11', 
    ce11_proms[1], 
    extension = 200, 
    window_size = 100,
    step_size = 10, 
    smooth_track = 1,
    motif = 'WW',
    period = 10,
    BPPARAM = setUpBPPARAM(1)
)
#> 
#>     The genome has been divided in 100-bp long windows 
#>     with a sliding window of 10-bp.
#>     29 windows overlap the 1 input loci (extended to 200-bp).
#>     The mapping will be split into 1 cores. Each core will 
#>     process 29 windows.
#>     
#>     Generating the following track: BSgenome.Celegans.UCSC.ce11_WW_10-bp-periodicity_g-100^10_smooth-1.bw
#>         GENOME:   BSgenome.Celegans.UCSC.ce11
#>         MOTIF:    WW
#>         PERIOD:   10
#>         # LOCI:   1
#>         # BASES:  380
#>     
#>     
#>     Now starting [2021-04-19 08:54:43]
#>     
#>     Merging the results...
#>     
#> 
#>    SUCCESS: BSgenome.Celegans.UCSC.ce11_WW_10-bp-periodicity_g-100^10_smooth-1.bw has been created!
#>    290 bases covered by the generated track.
#> 
#>     Finished without errors [2021-04-19 08:54:46]
#>     
track
#> RleList of length 7
#> $chrI
#> numeric-Rle of length 15072434 with 31 runs
#>   Lengths:       11202          10          10 ...          10    15060942
#>   Values : 0.00000e+00 2.14680e+00 1.15040e+00 ... 8.78000e-02 8.74301e-16
#> 
#> $chrII
#> numeric-Rle of length 15279421 with 1 run
#>   Lengths: 15279421
#>   Values :        0
#> 
#> $chrIII
#> numeric-Rle of length 13783801 with 1 run
#>   Lengths: 13783801
#>   Values :        0
#> 
#> $chrIV
#> numeric-Rle of length 17493829 with 1 run
#>   Lengths: 17493829
#>   Values :        0
#> 
#> $chrV
#> numeric-Rle of length 20924180 with 1 run
#>   Lengths: 20924180
#>   Values :        0
#> 
#> ...
#> <2 more elements>
unlink(
    'BSgenome.Celegans.UCSC.ce11_WW_10-bp-periodicity_g-100^10_smooth-1.bw'
)