GenomicFiles-class: GenomicFiles objects
In Bioconductor/GenomicFiles: Distributed computing by file or by range

GenomicFiles

R Documentation

GenomicFiles objects

Description

The GenomicFiles class is a matrix-like container where rows represent ranges of interest and columns represent files. The class is designed for byFile or byRange queries.

Constructor

: GenomicFiles(rowRanges, files, colData=DataFrame(), metadata=list(), ...):

Details

GenomicFiles inherits from the RangedSummarizedExperiment class in the SummarizedExperiment package. Currently, no use is made of the elementMetadat and assays slots. This may change in the future.

Accessors

In the code below, x is a GenomicFiles object.

rowRanges, rowRanges(x) <- value: Get or set the rowRanges on x. value can be a GRanges or GRangesList representing ranges or indices defined on the spaces (position) of the files.
files(x), files(x) <- value: Get or set the files on x. value can be a character() of file paths or a List of file objects such as BamFile, BigWigFile, FaFile, etc.
colData, colData(x) <- value: Get or set the colData on x. value must be a DataFrame instance describing the files. The number of rows must match the number of files. Row names, if present, become the column names of the GenomicFiles.
metadata, metadata(x) <- value: Get or set the metadata on x. value must be a SimpleList of arbitrary content describing the overall experiment.
dimnames, dimnames(x) <- value: Get or set the row and column names on x.

Methods

In the code below, x is a GenomicFiles object.

[: Subset the object by fileRange or fileSample.
show: Compactly display the object.
reduceByFile: Extract, manipulate and combine data defined in rowRanges within the files specified in files. See ?reduceByFile for details.
reduceByRange: Extract, manipulate and combine data defined in rowRanges across the files specified in files. See ?reduceByRange for details.

Author(s)

Martin Morgan and Valerie Obenchain

Examples

## -----------------------------------------------------------------------
## Basic Use
## -----------------------------------------------------------------------

if (require(RNAseqData.HNRNPC.bam.chr14)) { 
  fl <- RNAseqData.HNRNPC.bam.chr14_BAMFILES
  rd <- GRanges("chr14", 
                 IRanges(c(62262735, 63121531, 63980327), width=214700))
  cd <- DataFrame(method=rep("RNASeq", length(fl)),
                  format=rep("bam", length(fl)))

  ## Construct an instance of the class:
  gf <- GenomicFiles(files = fl, rowRanges = rd, colData = cd) 
  gf

  ## Subset on ranges or files for different experimental runs.
  dim(gf)
  gf_sub <- gf[2, 3:4]
  dim(gf_sub)
 
  ## When summarize = TRUE and no REDUCE is provided the reduceBy* 
  ## functions output a SummarizedExperiment object.
  MAP <- function(range, file, ...) {
      requireNamespace("GenomicFiles", quietly=TRUE) ## for coverage()
      requireNamespace("Rsamtools", quietly=TRUE)     ## for ScanBamParam()
      param = Rsamtools::ScanBamParam(which=range)
      GenomicFiles::coverage(file, param=param)[range]
  } 
  se <- reduceByRange(gf, MAP=MAP, summarize=TRUE)
  se
 
  ## Data from the rowRanges, colData and metadata slots in the
  ## GenomicFiles are transferred to the SummarizedExperiment.
  colData(se)
 
  ## Results are in the assays slot.
  assays(se) 
}
 
## -----------------------------------------------------------------------
## Managing cached or remote files with GenomicFiles
## -----------------------------------------------------------------------

## The GenomicFiles class can manage cached or remote files and their 
## associated ranges.

## Not run: 
## Files from AnnotationHub can be downloaded and cached locally.
library(AnnotationHub)
hub = AnnotationHub()
hublet = query(hub, c("files I'm", "interested in"))
# cache (if need) and return local path to files
fls = cache(hublet)

## An alternative to the local file paths is to use urls to a remote file.
## This approach could be used with something like rtracklayer::bigWig which
## supports remote file queries.
urls = hublet$sourceurls

## Define ranges of interest and use GenomicFiles to manage.
rngs = GRanges("chr10", IRanges(c(100000, 200000), width=1))
gf = GenomicFiles(rngs, fls)

## As an example, one could create a matrix from data extracted
## across multiple BED files.
MAP = function(rng, fl) {
    requireNamespace("rtracklayer", quietly=TRUE)  ## import, BEDFile
    rtracklayer::import(rtracklayer::BEDFile(fl), which=rng)$name
}
REDUCE = unlist
xx = reduceFiles(gf, MAP=MAP, REDUCE=REDUCE)
mcols(rngs) = simplify2array(xx)

## Data and ranges can be stored in a SummarizedExperiment.
SummarizedExperiment(list(my=simplify2array(xx)), rowRanges=rngs)

## End(Not run)

Bioconductor/GenomicFiles documentation built on Oct. 31, 2024, 7:01 a.m.