bcbioRNASeq-class: 'bcbioRNASeq' Object and Constructor
In roryk/bcbioRnaseq: RNA-Seq Utilities

Description Usage Arguments Details Value Metadata DESeq2 Remote Data Note Author(s) See Also Examples

bcbioRNASeq is an S4 class that extends RangedSummarizedExperiment, and is designed to store a bcbio RNA-seq analysis.

bcbioRNASeq(uploadDir, level = c("genes", "transcripts"),
  caller = c("salmon", "kallisto", "sailfish"), organism, samples = NULL,
  sampleMetadataFile = NULL, interestingGroups = "sampleName",
  ensemblRelease = NULL, genomeBuild = NULL, transgeneNames = NULL,
  spikeNames = NULL, gffFile = NULL, transformationLimit = 50L, ...)

`uploadDir`	Path to final upload directory. This path is set when running `bcbio_nextgen -w template`.
`level`	Import counts as "`genes`" (default) or "`transcripts`".
`caller`	Expression caller. Supports "`salmon`" (default), "`kallisto`", or "`sailfish`".
`organism`	Organism name. Use the full latin name (e.g. "Homo sapiens"), since this will be input downstream to AnnotationHub and ensembldb, unless `gffFile` is set. If set `NULL` (advanced use; not recommended), the function call will skip loading gene/transcript-level annotations into `rowRanges()`. This can be useful for poorly annotation genomes or experiments involving multiple genomes.
`samples`	Optional. Specify a subset of samples to load. The names must match the `description` specified in the bcbio YAML metadata. If a `sampleMetadataFile` is provided, that will take priority for sample selection. Typically this can be left unset.
`sampleMetadataFile`	Optional. Custom metadata file containing sample information. Otherwise defaults to sample metadata saved in the YAML file. Remote URLs are supported. Typically this can be left unset.
`interestingGroups`	Character vector denoting groups of interest that define the samples. If left unset, defaults to `sampleName`.
`ensemblRelease`	Optional. Ensembl release version. If unset, defaults to current release, and does not typically need to be user-defined. Passed to AnnotationHub for `EnsDb` annotation matching, unless `gffFile` is set.
`genomeBuild`	Optional. Ensembl genome build name (e.g. "GRCh38"). This will be passed to AnnotationHub for `EnsDb` annotation matching, unless `gffFile` is set.
`transgeneNames`	`character` vector indicating which `assay()` rows denote transgenes (e.g. EGFP, TDTOMATO).
`spikeNames`	`character` vector indicating which `assay()` rows denote spike-in sequences (e.g. ERCCs).
`gffFile`	Advanced use; not recommended. By default, we recommend leaving this `NULL` for genomes that are supported on Ensembl. In this case, the row annotations (`rowRanges()`) will be obtained automatically from Ensembl by passing the `organism`, `genomeBuild`, and `ensemblRelease` arguments to AnnotationHub and ensembldb. For a genome that is not supported on Ensembl and/or AnnotationHub, a GFF/GTF (General Feature Format) file is required. Generally, we recommend using a GTF (GFFv2) file here over a GFF3 file if possible, although all GFF formats are supported. The function will internally generate a `TxDb` containing transcript-to-gene mappings and construct a `GRanges` object containing the genomic ranges (`rowRanges()`).
`transformationLimit`	Maximum number of samples to calculate `DESeq2::rlog()` and `DESeq2::varianceStabilizingTransformation()` matrix. For large datasets, DESeq2 is slow to apply variance stabilization. In this case, we recommend loading up the dataset in a high-performance computing environment. Use `Inf` to always apply and `-Inf` to always skip.
`...`	Additional arguments, slotted into the `metadata()` accessor.

Simply point to the final upload directory generated by bcbio, and this constructor function will take care of the rest. It automatically imports RNA-seq counts, metadata, and the program versions used.

This class contains raw read counts and length-scaled transcripts per million (TPM) generated by tximport::tximport(). Counts can be loaded at gene or transcript level.

bcbioRNASeq.

The metadata() accessor contains:

Sample quality control metrics.
Ensembl annotations.
Server run paths.
R session information (e.g. utils::sessionInfo()).

DESeq2 is run automatically when bcbioRNASeq() is called, and variance stabilized counts are slotted into assays(). If the number of samples is bigger than the transformationLimit argument, rlog and vst counts will not be slotted into assays(). In this case, we recommend visualization using tmm() counts, which are automatically calculated using edgeR.

When working in RStudio, we recommend connecting to the bcbio run directory as a remote connection over sshfs.

bcbioRNASeq extended SummarizedExperiment prior to v0.2.0, where we migrated to RangedSummarizedExperiment.

Michael Steinbaugh, Lorena Pantano, Rory Kirchner, Victor Barrera

SummarizedExperiment::SummarizedExperiment().
methods::initialize().
methods::validObject().
BiocGenerics::updateObject().
.S4methods(class = "bcbioRNASeq").

Other S4 Class Definition: [,bcbioRNASeq,ANY,ANY,ANY-method, coerce, show, updateObject

uploadDir <- system.file("extdata/bcbio", package = "bcbioRNASeq")

# Gene level
x <- bcbioRNASeq(
    uploadDir = uploadDir,
    level = "genes",
    caller = "salmon",
    organism = "Mus musculus",
    ensemblRelease = 87L
)
show(x)
is(x, "RangedSummarizedExperiment")
validObject(x)

# Transcript level
x <- bcbioRNASeq(
    uploadDir = uploadDir,
    level = "transcripts",
    caller = "salmon",
    organism = "Mus musculus",
    ensemblRelease = 87L
)
show(x)
validObject(x)