TrackHub Genome Access

Description

A TrackHub data source is a collection of tracks and sequences, separated by genome. This class, TrackHubGenome provides direct access to the data for one particular genome.

Constructor

TrackHubGenome(trackhub, genome, create = FALSE: Constructs a new TrackHubGenome object, representing genome in the repository trackhub (a URI string or a TrackHub object).

The genome argument can be an ID corresponding to a genome (potentially) in trackhub or an installed BSgenome package.

If create is TRUE, and the trackDb file does not already exist, it will be created. Creation only works if the repository is local and writeable.

Accessor Methods

In the code snippets below, x represent a TrackHubGenome object.

uri(x): Get the uri pointing to the genome directory in the TrackHub repository.

genome(x): Get the name of the genome, e.g. “hg19”.

length(x): number of tracks

names(x), trackNames(x): names of the tracks

getTracks(x): Get the List of Track from the tracks

trackhub(x): Get the TrackHub object that contains this genome.

organism(x): Get the organism name for this genome, e.g., “H sapiens”.

trackField(x, name, field): Get the value of field for name track.

trackField(x, name, field) <- value: Store the field and value for name track.

writeTrackHub(x): Write tracks from memory representation to the trackDb file.

Data Access

track(x, name), x$name: get the track called name

track(x, name, format = bestFileFormat(value)) <- value, x$name <- value: store the track value under name. Note that track storing is only supported for local repositories, i.e., those with a file:// URI scheme.

Currently, supported value types include a GenomicRanges, GRangesList, or a file resource (copied to the repository). The file resource may be represented as a path, URL, BiocFile or RsamtoolsFile. If not a file name, value is written in format. For generic interval data, this means a BigWig file (if there is a numeric “score” column) or a BED file otherwise. An RleList (e.g., coverage) is output as BigWig. For UCSCData values, the format is chosen according to the type of track line. For RsamtoolsFile objects, the file and its index are copied.

referenceSequence(x): Get the reference sequence, as a DNAStringSet.

referenceSequence(x) <- value: Set the reference sequence, as a DNAStringSet. It is written as a 2bit file. This only works on local repositories.

Author(s)

Michael Lawrence

Examples

tests_dir <- system.file("tests", package = "rtracklayer")
th <- TrackHub(file.path(tests_dir, "trackhub"))
thg <- TrackHubGenome(th, "hg19")
length(thg)
organism(thg)
names(thg)

## Not run: 
th <- TrackHub(file.path(tests_dir, "trackhub"), create = TRUE)
genomesFile(th) <- "genomes.txt"
genomeInfo(th) <- Genome(genome = "hg38", trackDb = "hg38/trackDb.txt")
genomeField(th, "hg38", "twoBitPath") <- "hg38/seq.2bit"
writeTrackHub(th)
thg <- TrackHubGenome(th, "hg38", create = TRUE)
seq <- import(file.path(tests_dir, "test.2bit"))
referenceSequence(thg) <- seq
track(thg, "PeaksData") <- paste0(tests_dir, "/test.bigWig")
trackField(thg, "wgEncodeUWDukeDnaseGM12878FdrPeaks", "bigDataUrl") <- "hg38/wgEncodeCshlShortRnaSeq.bigWig"
trackField(thg, "wgEncodeUWDukeDnaseGM12878FdrPeaks", "color") <- "8,104,172"
writeTrackHub(thg)

## End(Not run)