Extract non-overlapping exonic or intronic parts from a TxDb-like object

Description

exonicParts and intronicParts extract the non-overlapping (a.k.a. disjoint) exonic or intronic parts from a TxDb-like object.

Usage

exonicParts(txdb, linked.to.single.gene.only=FALSE)
intronicParts(txdb, linked.to.single.gene.only=FALSE)

Arguments

Value

exonicParts returns a disjoint and strictly sorted GRanges object with 1 range per exonic part and with metadata columns tx_id, tx_name, gene_id, exon_id, exon_name, and exon_rank.

intronicParts returns a disjoint and strictly sorted GRanges object with 1 range per intronic part and with metadata columns tx_id, tx_name, and gene_id.

Note

exonicParts is a replacement for disjointExons with the following differences/improvements:

• Argument linked.to.single.gene.only in exonicParts replaces argument aggregateGenes in disjointExons, but has opposite meaning i.e. exonicParts(txdb, linked.to.single.gene.only=TRUE) returns the same exonic parts as disjointExons(txdb, aggregateGenes=FALSE).

• Unlike disjointExons(txdb, aggregateGenes=TRUE), exonicParts(txdb, linked.to.single.gene.only=FALSE) does NOT discard exon parts that are not linked to a gene.

• exonicParts is almost 2x more efficient than disjointExons.

• exonicParts works out-of-the-box on any TxDb-like object that supports the transcripts() and exonsBy() extractors (e.g. on an EnsDb object).

Author(s)

Hervé Pagès

See Also

• disjoin in the IRanges package.

• transcripts, transcriptsBy, and transcriptsByOverlaps, for extracting genomic feature locations from a TxDb-like object.

• transcriptLengths for extracting the transcript lengths (and other metrics) from a TxDb object.

• extractTranscriptSeqs for extracting transcript (or CDS) sequences from chromosome sequences.

• coverageByTranscript for computing coverage by transcript (or CDS) of a set of ranges.

• The TxDb class.

Examples

library(TxDb.Hsapiens.UCSC.hg19.knownGene)
txdb <- TxDb.Hsapiens.UCSC.hg19.knownGene

## ---------------------------------------------------------------------
## exonicParts()
## ---------------------------------------------------------------------

exonic_parts1 <- exonicParts(txdb)
exonic_parts1

## Mapping from exonic parts to genes is many-to-many:
mcols(exonic_parts1)$gene_id
table(lengths(mcols(exonic_parts1)$gene_id))
## A Human exonic part can be linked to 0 to 22 known genes!

exonic_parts2 <- exonicParts(txdb, linked.to.single.gene.only=TRUE)
exonic_parts2

## Mapping from exonic parts to genes now is many-to-one:
class(mcols(exonic_parts2)$gene_id)

## Sanity checks:
stopifnot(isDisjoint(exonic_parts1), isStrictlySorted(exonic_parts1))
stopifnot(isDisjoint(exonic_parts2), isStrictlySorted(exonic_parts2))
stopifnot(all(exonic_parts2 %within% reduce(exonic_parts1)))
stopifnot(identical(
    lengths(mcols(exonic_parts1)$gene_id) == 1L,
    exonic_parts1 %within% exonic_parts2
))

## ---------------------------------------------------------------------
## intronicParts()
## ---------------------------------------------------------------------

intronic_parts1 <- intronicParts(txdb)
intronic_parts1

## Mapping from intronic parts to genes is many-to-many:
mcols(intronic_parts1)$gene_id
table(lengths(mcols(intronic_parts1)$gene_id))
## A Human intronic part can be linked to 0 to 22 known genes!

intronic_parts2 <- intronicParts(txdb, linked.to.single.gene.only=TRUE)
intronic_parts2

## Mapping from intronic parts to genes now is many-to-one:
class(mcols(intronic_parts2)$gene_id)

## Sanity checks:
stopifnot(isDisjoint(intronic_parts1), isStrictlySorted(intronic_parts1))
stopifnot(isDisjoint(intronic_parts2), isStrictlySorted(intronic_parts2))
stopifnot(all(intronic_parts2 %within% reduce(intronic_parts1)))
stopifnot(identical(
    lengths(mcols(intronic_parts1)$gene_id) == 1L,
    intronic_parts1 %within% intronic_parts2
))