Retrieve annotation data from an Ensembl based package
Retrieve gene/transcript/exons annotations stored in an Ensembl based
database package generated with the makeEnsembldbPackage
function. Parameter filter enables to define filters to
retrieve only specific data. Alternatively, a global filter might be
added to the EnsDb object using the addFilter
method.
## S4 method for signature 'EnsDb'
exons(x, columns = listColumns(x,"exon"),
filter = AnnotationFilterList(), order.by,
order.type = "asc", return.type = "GRanges")
## S4 method for signature 'EnsDb'
exonsBy(x, by = c("tx", "gene"),
columns = listColumns(x, "exon"), filter =
AnnotationFilterList(), use.names = FALSE)
## S4 method for signature 'EnsDb'
intronsByTranscript(x, ..., use.names = FALSE)
## S4 method for signature 'EnsDb'
exonsByOverlaps(x, ranges, maxgap = -1L, minoverlap = 0L,
type = c("any", "start", "end"), columns = listColumns(x, "exon"),
filter = AnnotationFilterList())
## S4 method for signature 'EnsDb'
transcripts(x, columns = listColumns(x, "tx"),
filter = AnnotationFilterList(), order.by, order.type = "asc",
return.type = "GRanges")
## S4 method for signature 'EnsDb'
transcriptsBy(x, by = c("gene", "exon"),
columns = listColumns(x, "tx"), filter = AnnotationFilterList())
## S4 method for signature 'EnsDb'
transcriptsByOverlaps(x, ranges, maxgap = -1L,
minoverlap = 0L, type = c("any", "start", "end"),
columns = listColumns(x, "tx"), filter = AnnotationFilterList())
## S4 method for signature 'EnsDb'
promoters(x, upstream = 2000, downstream = 200,
use.names = TRUE, ...)
## S4 method for signature 'EnsDb'
genes(x, columns = c(listColumns(x, "gene"), "entrezid"),
filter = AnnotationFilterList(), order.by, order.type = "asc",
return.type = "GRanges")
## S4 method for signature 'EnsDb'
disjointExons(x, aggregateGenes = FALSE,
includeTranscripts = TRUE, filter = AnnotationFilterList(), ...)
## S4 method for signature 'EnsDb'
cdsBy(x, by = c("tx", "gene"), columns = NULL,
filter = AnnotationFilterList(), use.names = FALSE)
## S4 method for signature 'EnsDb'
fiveUTRsByTranscript(x, columns = NULL,
filter = AnnotationFilterList())
## S4 method for signature 'EnsDb'
threeUTRsByTranscript(x, columns = NULL,
filter = AnnotationFilterList())
## S4 method for signature 'GRangesList'
toSAF(x, ...)(In alphabetic order)
... |
For |
aggregateGenes |
For |
by |
For |
columns |
Columns to be retrieved from the database tables. Default values for Note that any of the column names of the database tables can be
submitted to any of the methods (use For |
downstream |
For method |
filter |
A filter describing which results to retrieve from the database. Can
be a single object extending
|
includeTranscripts |
For |
maxgap |
For |
minoverlap |
For |
order.by |
Character vector specifying the column(s) by which the result should
be ordered. This can be either in the form of
|
order.type |
If the results should be ordered ascending
( |
ranges |
For |
return.type |
Type of the returned object. Can be either
|
type |
For |
upstream |
For method |
use.names |
For |
x |
For |
A detailed description of all database tables and the associated attributes/column names is also given in the vignette of this package. An overview of the columns is given below:
the Ensembl gene ID of the gene.
the name of the gene (in most cases its official symbol).
the NCBI Entrezgene ID of the gene. Note that this
column contains a list of Entrezgene identifiers to
accommodate the potential 1:n mapping between Ensembl genes and
Entrezgene IDs.
the biotype of the gene.
the start coordinate of the gene on the sequence (usually a chromosome).
the end coordinate of the gene.
the name of the sequence the gene is encoded (usually a chromosome).
the strand on which the gene is encoded
the coordinate system of the sequence.
the Ensembl transcript ID.
the biotype of the transcript.
the chromosomal start coordinate of the transcript.
the chromosomal end coordinate of the transcript.
the start coordinate of the coding region of the transcript (NULL for non-coding transcripts).
the end coordinate of the coding region.
the G and C nucleotide content of the transcript's sequence expressed as a percentage (i.e. between 0 and 100).
the ID of the exon. In Ensembl, each exon specified by a unique chromosomal start and end position has its own ID. Thus, the same exon might be part of several transcripts.
the chromosomal start coordinate of the exon.
the chromosomal end coordinate of the exon.
the index of the exon in the transcript model. As noted above, an exon can be part of several transcripts and thus its position inside these transcript might differ.
Many EnsDb databases provide also protein related
annotations. See listProteinColumns for more information.
For exons, transcripts and genes,
a data.frame, DataFrame
or a GRanges, depending on the value of the
return.type parameter. The result is ordered as specified by
the parameter order.by or, if not provided, by seq_name
and chromosomal start coordinate, but NOT by any ordering of values in
eventually submitted filter objects.
For exonsBy, transcriptsBy:
a GRangesList, depending on the value of the
return.type parameter. The results are ordered by the value of the
by parameter.
For exonsByOverlaps and transcriptsByOverlaps: a
GRanges with the exons or transcripts overlapping the specified
regions.
For toSAF: a data.frame with column names
"GeneID" (the group name from the GRangesList, i.e. the
ID by which the GRanges are split), "Chr" (the seqnames
from the GRanges), "Start" (the start coordinate),
"End" (the end coordinate) and "Strand" (the strand).
For disjointExons: a GRanges of non-overlapping exon
parts.
For cdsBy: a GRangesList with GRanges per either
transcript or exon specifying the start and end coordinates of the
coding region of the transcript or gene.
For fiveUTRsByTranscript: a GRangesList with
GRanges for each protein coding transcript representing the
start and end coordinates of full or partial exons that constitute the
5' untranslated region of the transcript.
For threeUTRsByTranscript: a GRangesList with
GRanges for each protein coding transcript representing the
start and end coordinates of full or partial exons that constitute the
3' untranslated region of the transcript.
Retrieve exon information from the database. Additional columns from transcripts or genes associated with the exons can be specified and are added to the respective exon annotation.
Retrieve exons grouped by transcript or by gene. This
function returns a GRangesList as does the analogous function
in the GenomicFeatures package. Using the columns
parameter it is possible to determine which additional values should
be retrieved from the database. These will be included in the
GRanges object for the exons as metadata columns.
The exons in the inner GRanges are ordered by the exon
index within the transcript (if by="tx"), or increasingly by the
chromosomal start position of the exon or decreasingly by the chromosomal end
position of the exon depending whether the gene is encoded on the
+ or - strand (for by="gene").
The GRanges in the GRangesList will be ordered by
the name of the gene or transcript.
Retrieve introns by transcripts. Filters can also be passed to the
function. For more information see the intronsByTranscript
method in the GenomicFeatures package.
Retrieve exons overlapping specified genomic ranges. For
more information see the
exonsByOverlaps method in the
GenomicFeatures package. The functionality is to some
extent similar and redundant to the exons method in
combination with GRangesFilter filter.
Retrieve transcript information from the database. Additional columns from genes or exons associated with the transcripts can be specified and are added to the respective transcript annotation.
Retrieve transcripts grouped by gene or exon. This
function returns a GRangesList as does the analogous function
in the GenomicFeatures package. Using the columns
parameter it is possible to determine which additional values should
be retrieved from the database. These will be included in the
GRanges object for the transcripts as metadata columns.
The transcripts in the inner GRanges are ordered increasingly by the
chromosomal start position of the transcript for genes encoded on
the + strand and in a decreasing manner by the chromosomal end
position of the transcript for genes encoded on the - strand.
The GRanges in the GRangesList will be ordered by
the name of the gene or exon.
Retrieve transcripts overlapping specified genomic ranges. For
more information see
transcriptsByOverlaps method in the
GenomicFeatures package. The functionality is to some
extent similar and redundant to the transcripts method in
combination with GRangesFilter filter.
Retrieve promoter information from the database. Additional columns from genes or exons associated with the promoters can be specified and are added to the respective promoter annotation.
Retrieve gene information from the database. Additional columns
from transcripts or exons associated with the genes can be
specified and are added to the respective gene annotation. Note
that column "entrezid" is a list of Entrezgene
identifiers to accomodate the potential 1:n mapping between
Ensembl genes and Entrezgene IDs.
This method is identical to
disjointExons defined in the
GenomicFeatures package. It creates a GRanges of
non-overlapping exon parts with metadata columns of gene_id
and exonic_part. Exon parts that overlap more than one gene
can be dropped with aggregateGenes=FALSE.
Returns the coding region grouped either by transcript or by
gene. Each element in the GRangesList represents the cds
for one transcript or gene, with the individual ranges
corresponding to the coding part of its exons.
For by="tx" additional annotation columns can be added to
the individual GRanges (in addition to the default columns
exon_id and exon_rank).
Note that the GRangesList is sorted by its names.
Returns the 5' untranslated region for protein coding transcripts.
Returns the 3' untranslated region for protein coding transcripts.
Reformats a GRangesList object into a
data.frame corresponding to a standard SAF (Simplified
Annotation Format) file (i.e. with column names "GeneID",
"Chr", "Start", "End" and
"Strand"). Note: this method makes only sense on a
GRangesList that groups features (exons, transcripts) by gene.
Ensembl defines genes not only on standard chromosomes, but also on
patched chromosomes and chromosome variants. Thus it might be
advisable to restrict the queries to just those chromosomes of
interest (e.g. by specifying a SeqNameFilter(c(1:22, "X", "Y"))).
In addition, also so called LRG genes (Locus Reference Genomic) are defined in
Ensembl. Their gene id starts with LRG instead of ENS for Ensembl
genes, thus, a filter can be applied to specifically select those
genes or exclude those genes (see examples below).
Depending on the value of the global option
"ucscChromosomeNames" (use
getOption(ucscChromosomeNames, FALSE) to get its value or
option(ucscChromosomeNames=TRUE) to change its value)
the sequence/chromosome names of the returned GRanges objects
or provided in the returned data.frame or DataFrame
correspond to Ensembl chromosome names (if value is FALSE) or
UCSC chromosome names (if TRUE). This ensures a better
integration with the Gviz package, in which this option is set
by default to TRUE.
While it is possible to request values from a column "tx_name"
(with the columns argument), no such column is present in the
database. The returned values correspond to the ID of the transcripts.
Johannes Rainer, Tim Triche
supportedFilters to get an overview of supported filters.
makeEnsembldbPackage,
listColumns, lengthOf
addFilter for globally adding filters to an EnsDb
object.
library(EnsDb.Hsapiens.v86)
edb <- EnsDb.Hsapiens.v86
###### genes
##
## Get all genes encoded on chromosome Y
AllY <- genes(edb, filter = SeqNameFilter("Y"))
AllY
## Return the result as a DataFrame; also, we use a filter expression here
## to define which features to extract from the database.
AllY.granges <- genes(edb,
filter = ~ seq_name == "Y",
return.type="DataFrame")
AllY.granges
## Include all transcripts of the gene and their chromosomal
## coordinates, sort by chrom start of transcripts and return as
## GRanges.
AllY.granges.tx <- genes(edb,
filter = SeqNameFilter("Y"),
columns = c("gene_id", "seq_name",
"seq_strand", "tx_id", "tx_biotype",
"tx_seq_start", "tx_seq_end"),
order.by = "tx_seq_start")
AllY.granges.tx
###### transcripts
##
## Get all transcripts of a gene
Tx <- transcripts(edb,
filter = GeneIdFilter("ENSG00000184895"),
order.by = "tx_seq_start")
Tx
## Get all transcripts of two genes along with some information on the
## gene and transcript
Tx <- transcripts(edb,
filter = GeneIdFilter(c("ENSG00000184895",
"ENSG00000092377")),
columns = c("gene_id", "gene_seq_start", "gene_seq_end",
"gene_biotype", "tx_biotype"))
Tx
###### promoters
##
## Get the bona-fide promoters (2k up- to 200nt downstream of TSS)
promoters(edb, filter = GeneIdFilter(c("ENSG00000184895",
"ENSG00000092377")))
###### exons
##
## Get all exons of protein coding transcript for the gene ENSG00000184895
Exon <- exons(edb,
filter = ~ gene_id == "ENSG00000184895" &
tx_biotype == "protein_coding",
columns = c("gene_id", "gene_seq_start", "gene_seq_end",
"tx_biotype", "gene_biotype"))
Exon
##### exonsBy
##
## Get all exons for transcripts encoded on chromosomes X and Y.
ETx <- exonsBy(edb, by = "tx",
filter = SeqNameFilter(c("X", "Y")))
ETx
## Get all exons for genes encoded on chromosome 1 to 22, X and Y and
## include additional annotation columns in the result
EGenes <- exonsBy(edb, by = "gene",
filter = SeqNameFilter(c("X", "Y")),
columns = c("gene_biotype", "gene_name"))
EGenes
## Note that this might also contain "LRG" genes.
length(grep(names(EGenes), pattern="LRG"))
## to fetch just Ensemblgenes, use an GeneIdFilter with value
## "ENS%" and condition "like"
eg <- exonsBy(edb, by = "gene",
filter = AnnotationFilterList(SeqNameFilter(c("X", "Y")),
GeneIdFilter("ENS", "startsWith")),
columns = c("gene_biotype", "gene_name"))
eg
length(grep(names(eg), pattern="LRG"))
##### transcriptsBy
##
TGenes <- transcriptsBy(edb, by = "gene",
filter = SeqNameFilter(c("X", "Y")))
TGenes
## convert this to a SAF formatted data.frame that can be used by the
## featureCounts function from the Rsubreader package.
head(toSAF(TGenes))
##### transcriptsByOverlaps
##
ir <- IRanges(start = c(2654890, 2709520, 28111770),
end = c(2654900, 2709550, 28111790))
gr <- GRanges(rep("Y", length(ir)), ir)
## Retrieve all transcripts overlapping any of the regions.
txs <- transcriptsByOverlaps(edb, gr)
txs
## Alternatively, use a GRangesFilter
grf <- GRangesFilter(gr, type = "any")
txs <- transcripts(edb, filter = grf)
txs
#### cdsBy
## Get the coding region for all transcripts on chromosome Y.
## Specifying also additional annotation columns (in addition to the default
## exon_id and exon_rank).
cds <- cdsBy(edb, by = "tx", filter = SeqNameFilter("Y"),
columns = c("tx_biotype", "gene_name"))
#### the 5' untranslated regions:
fUTRs <- fiveUTRsByTranscript(edb, filter = SeqNameFilter("Y"))
#### the 3' untranslated regions with additional column gene_name.
tUTRs <- threeUTRsByTranscript(edb, filter = SeqNameFilter("Y"),
columns = "gene_name")Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.