ScoreMatrixList-methods: Make ScoreMatrixList from multiple targets

Description Usage Arguments Value Examples

Description

The function constructs a list of ScoreMatrix objects in the form of ScoreMatrixList object. This object can be visualized using multiHeatMatrix, heatMeta or plotMeta

Usage

1
2
3
4
ScoreMatrixList(targets, windows = NULL, bin.num = NULL, bin.op = "mean",
  strand.aware = FALSE, weight.col = NULL, is.noCovNA = FALSE,
  type = "auto", rpm = FALSE, unique = FALSE, extend = 0,
  param = NULL, library.size = NULL, cores = 1)

Arguments

targets

can be a list of scoreMatrix objects, that are coerced to the ScoreMatrixList, a list of RleList objects, or a character vector specifying the locations of mulitple bam files or bigWig files that are used to construct the scoreMatrixList. If it is either a RleList object or a character vector of files, it is obligatory to give a windows argument.

windows

GenomicRanges containing viewpoints for the scoreMatrix or ScoreMatrixList functions

bin.num

an integer telling the number of bins to bin the score matrix

bin.op

an name of the function that will be used for smoothing windows of ranges

strand.aware

a boolean telling the function whether to reverse the coverage of ranges that come from - strand (e.g. when plotting enrichment around transcription start sites)

weight.col

if the object is GRanges object a numeric column in meta data part can be used as weights. This is particularly useful when genomic regions have scores other than their coverage values, such as percent methylation, conservation scores, GC content, etc.

is.noCovNA

(Default:FALSE) if TRUE,and if 'targets' is a GRanges object with 'weight.col' provided, the bases that are uncovered will be preserved as NA in the returned object. This useful for situations where you can not have coverage all over the genome, such as CpG methylation values.

type

(Default:"auto") if targets is a character vector of file paths, then type designates the type of the corresponding files (bam or bigWig)

rpm

boolean telling whether to normalize the coverage to per milion reads. FALSE by default. See library.size.

unique

boolean which tells the function to remove duplicated reads based on chr, start, end and strand

extend

numeric which tells the function to extend the features ( i.e aligned reads) to total length ofwidth+extend

param

ScanBamParam object

library.size

a numeric vector of the same length as targets indicating total number of mapped reads in BAM files (targets). If is not given (default: NULL) then library sizes for every target is calculated using the Rsamtools idxstatsBam function: sum(idxstatsBam(target)$mapped). rpm argument has to be set to TRUE.

cores

the number of cores to use (default: 1)

Value

returns a ScoreMatrixList object

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
# visualize the distribution of cage clusters and cpg islands around promoters
library(GenomicRanges)
data(cage)
data(cpgi)
data(promoters)
 
cage$tpm = NULL
targets = GRangesList(cage=cage, cpgi=cpgi)
sml = ScoreMatrixList(targets, promoters, bin.num=10, strand.aware=TRUE)
sml

multiHeatMatrix(sml)

BIMSBbioinfo/genomation documentation built on Feb. 4, 2018, 12:11 p.m.