Generating an using Ensembl based annotation packages
6 May 2018
Package
ensembldb 2.4.1
1 Introduction
The ensembldb package provides functions to create and use transcript centric
annotation databases/packages. The annotation for the databases are directly
fetched from Ensembl 1 using their Perl API. The functionality and data is
similar to that of the TxDb packages from the GenomicFeatures package, but, in
addition to retrieve all gene/transcript models and annotations from the
database, the ensembldb package provides also a filter framework allowing to
retrieve annotations for specific entries like genes encoded on a chromosome
region or transcript models of lincRNA genes. From version 1.7 on, EnsDb
databases created by the ensembldb package contain also protein annotation data
(see Section 11 for the database layout and an overview of
available attributes/columns). For more information on the use of the protein
annotations refer to the proteins vignette.
Another main goal of this package is to generate versioned annotation
packages, i.e. annotation packages that are build for a specific Ensembl
release, and are also named according to that (e.g. EnsDb.Hsapiens.v86 for
human gene definitions of the Ensembl code database version 86). This ensures
reproducibility, as it allows to load annotations from a specific Ensembl
release also if newer versions of annotation packages/releases are available. It
also allows to load multiple annotation packages at the same time in order to
e.g. compare gene models between Ensembl releases.
In the example below we load an Ensembl based annotation package for Homo
sapiens, Ensembl version 86. The EnsDb object providing access to the underlying
SQLite database is bound to the variable name EnsDb.Hsapiens.v86.
library(EnsDb.Hsapiens.v86)
## Making a "short cut"
edb <- EnsDb.Hsapiens.v86
## print some informations for this package
edb## EnsDb for Ensembl:
## |Backend: SQLite
## |Db type: EnsDb
## |Type of Gene ID: Ensembl Gene ID
## |Supporting package: ensembldb
## |Db created by: ensembldb package from Bioconductor
## |script_version: 0.3.0
## |Creation time: Thu May 18 16:32:27 2017
## |ensembl_version: 86
## |ensembl_host: localhost
## |Organism: homo_sapiens
## |taxonomy_id: 9606
## |genome_build: GRCh38
## |DBSCHEMAVERSION: 2.0
## | No. of genes: 63970.
## | No. of transcripts: 216741.
## |Protein data available.## For what organism was the database generated?
organism(edb)## [1] "Homo sapiens"2 Using ensembldb annotation packages to retrieve specific annotations
One of the strengths of the ensembldb package and the related EnsDb databases is
its implementation of a filter framework that enables to efficiently extract
data sub-sets from the databases. The ensembldb package supports most of the
filters defined in the AnnotationFilter Bioconductor package and defines some
additional filters specific to the data stored in EnsDb databases. Filters can
be passed directly to all methods extracting data from an EnsDb (such as genes,
transcripts or exons). Alternatively it is possible with the addFilter or filter
functions to add a filter directly to an EnsDb which will then be used in all
queries on that object.
The supportedFilters method can be used to get an overview over all supported
filter classes, each of them (except the GRangesFilter) working on a single
column/field in the database.
supportedFilters(edb)## filter field
## 1 EntrezFilter entrez
## 2 ExonEndFilter exon_end
## 3 ExonIdFilter exon_id
## 4 ExonRankFilter exon_rank
## 5 ExonStartFilter exon_start
## 6 GRangesFilter <NA>
## 7 GeneBiotypeFilter gene_biotype
## 8 GeneEndFilter gene_end
## 9 GeneIdFilter gene_id
## 10 GeneStartFilter gene_start
## 11 GenenameFilter genename
## 12 ProtDomIdFilter prot_dom_id
## 13 ProteinDomainIdFilter protein_domain_id
## 14 ProteinDomainSourceFilter protein_domain_source
## 15 ProteinIdFilter protein_id
## 16 SeqNameFilter seq_name
## 17 SeqStrandFilter seq_strand
## 18 SymbolFilter symbol
## 19 TxBiotypeFilter tx_biotype
## 20 TxEndFilter tx_end
## 21 TxIdFilter tx_id
## 22 TxNameFilter tx_name
## 23 TxStartFilter tx_start
## 24 UniprotDbFilter uniprot_db
## 25 UniprotFilter uniprot
## 26 UniprotMappingTypeFilter uniprot_mapping_typeThese filters can be divided into 3 main filter types:
IntegerFilter: filter classes extending this basic object can take a single numeric value as input and support the conditions ==, !=, >, <, >= and <=. All filters that work on chromosomal coordinates, such as theGeneEndFilterextendIntegerFilter.CharacterFilter: filter classes extending this object can take a single or multiple character values as input and allow conditions: ==, !=, “startsWith” , “endsWith” and “contains”. All filters working on IDs extend this class.GRangesFilter: takes aGRangesobject as input and supports all conditions thatfindOverlapsfrom theIRangespackage supports (“any”, “start”, “end”, “within”, “equal”). Note that these have to be passed using the parametertypeto the constructor function.
The supported filters are:
EntrezFilter: allows to filter results based on NCBI Entrezgene identifiers of the genes.ExonEndFilter: filter using the chromosomal end coordinate of exons.ExonIdFilter: filter based on the (Ensembl) exon identifiers.ExonRankFilter: filter based on the rank (index) of an exon within the transcript model. Exons are always numbered from 5’ to 3’ end of the transcript, thus, also on the reverse strand, the exon 1 is the most 5’ exon of the transcript.ExonStartFilter: filter using the chromosomal start coordinate of exons.GeneBiotypeFilter: filter using the gene biotypes defined in the Ensembl database; use thelistGenebiotypesmethod to list all available biotypes.GeneEndFilter: filter using the chromosomal end coordinate of gene.GeneIdFilter: filter based on the Ensembl gene IDs.GenenameFilter: filter based on the names (symbols) of the genes.GeneStartFilter: filter using the chromosomal start coordinate of gene.GRangesFilter: allows to retrieve all features (genes, transcripts or exons) that are either within (setting parametertypeto “within”) or partially overlapping (settingtypeto “any”) the defined genomic region/range. Note that, depending on the called method (genes,transcriptsorexons) the start and end coordinates of either the genes, transcripts or exons are used for the filter. For methodsexonsBy,cdsByandtxBythe coordinates ofbyare used.SeqNameFilter: filter by the name of the chromosomes the genes are encoded on.SeqStrandFilter: filter for the chromosome strand on which the genes are encoded.SymbolFilter: filter on gene symbols; note that no database columns symbol is available in anEnsDbdatabase and hence the gene name is used for filtering.TxBiotypeFilter: filter on the transcript biotype defined in Ensembl; use thelistTxbiotypesmethod to list all available biotypes.TxEndFilter: filter using the chromosomal end coordinate of transcripts.TxIdFilter: filter on the Ensembl transcript identifiers.TxNameFilter: filter on the Ensembl transcript names (currently identical to the transcript IDs).TxStartFilter: filter using the chromosomal start coordinate of transcripts.
In addition to the above listed DNA-RNA-based filters, protein-specific filters are also available:
ProtDomIdFilter,ProteinDomainIdFilter: filter by the protein domain ID.ProteinDomainSourceFilter: filter by the source of the protein domain (database or method, e.g. pfam).ProteinIdFilter: filter by Ensembl protein ID filters.UniprotDbFilter: filter by the name of the Uniprot database.UniprotFilter: filter by the Uniprot ID.UniprotMappingTypeFilter: filter by the mapping type of Ensembl protein IDs to Uniprot IDs.
These can however only be used on EnsDb databases that provide protein
annotations, i.e. for which a call to hasProteinData returns TRUE.
EnsDb databases for more recent Ensembl versions (starting from Ensembl 87)
provide also evidence levels for individual transcripts in the tx_support_level
database column. Such databases support also a TxSupportLevelFilter filter to
use this columns for filtering.
A simple use case for the filter framework would be to get all transcripts for
the gene BCL2L11. To this end we specify a GenenameFilter with the value
BCL2L11. As a result we get a GRanges object with start, end, strand and seqname
being the start coordinate, end coordinate, chromosome name and strand for the
respective transcripts. All additional annotations are available as metadata
columns. Alternatively, by setting return.type to “DataFrame”, or “data.frame”
the method would return a DataFrame or data.frame object instead of the default
GRanges.
Tx <- transcripts(edb, filter = list(GenenameFilter("BCL2L11")))
Tx## GRanges object with 28 ranges and 7 metadata columns:
## seqnames ranges strand | tx_id
## <Rle> <IRanges> <Rle> | <character>
## ENST00000432179 2 111119378-111124112 + | ENST00000432179
## ENST00000308659 2 111120914-111165048 + | ENST00000308659
## ENST00000337565 2 111120914-111128844 + | ENST00000337565
## ENST00000622509 2 111120914-111168445 + | ENST00000622509
## ENST00000619294 2 111120914-111168445 + | ENST00000619294
## ... ... ... ... . ...
## ENST00000452231 2 111123746-111164231 + | ENST00000452231
## ENST00000361493 2 111123746-111164231 + | ENST00000361493
## ENST00000431217 2 111123746-111164352 + | ENST00000431217
## ENST00000439718 2 111123746-111164643 + | ENST00000439718
## ENST00000438054 2 111123752-111146284 + | ENST00000438054
## tx_biotype tx_cds_seq_start tx_cds_seq_end
## <character> <integer> <integer>
## ENST00000432179 protein_coding 111123746 111124112
## ENST00000308659 protein_coding 111123746 111164231
## ENST00000337565 protein_coding 111123746 111128751
## ENST00000622509 protein_coding 111123746 111161439
## ENST00000619294 protein_coding 111123746 111144501
## ... ... ... ...
## ENST00000452231 nonsense_mediated_decay 111123746 111161439
## ENST00000361493 nonsense_mediated_decay 111123746 111130235
## ENST00000431217 nonsense_mediated_decay 111123746 111144501
## ENST00000439718 nonsense_mediated_decay 111123746 111151851
## ENST00000438054 protein_coding 111123752 111144491
## gene_id tx_name gene_name
## <character> <character> <character>
## ENST00000432179 ENSG00000153094 ENST00000432179 BCL2L11
## ENST00000308659 ENSG00000153094 ENST00000308659 BCL2L11
## ENST00000337565 ENSG00000153094 ENST00000337565 BCL2L11
## ENST00000622509 ENSG00000153094 ENST00000622509 BCL2L11
## ENST00000619294 ENSG00000153094 ENST00000619294 BCL2L11
## ... ... ... ...
## ENST00000452231 ENSG00000153094 ENST00000452231 BCL2L11
## ENST00000361493 ENSG00000153094 ENST00000361493 BCL2L11
## ENST00000431217 ENSG00000153094 ENST00000431217 BCL2L11
## ENST00000439718 ENSG00000153094 ENST00000439718 BCL2L11
## ENST00000438054 ENSG00000153094 ENST00000438054 BCL2L11
## -------
## seqinfo: 1 sequence from GRCh38 genome## as this is a GRanges object we can access e.g. the start coordinates with
head(start(Tx))## [1] 111119378 111120914 111120914 111120914 111120914 111120914## or extract the biotype with
head(Tx$tx_biotype)## [1] "protein_coding" "protein_coding" "protein_coding" "protein_coding"
## [5] "protein_coding" "protein_coding"The parameter columns of the extractor methods (such as exons, genes or
transcripts) allows to specify which database attributes (columns) should be
retrieved. The exons method returns by default all exon-related columns, the
transcripts all columns from the transcript database table and the genes all
from the gene table. Note however that in the example above we got also a column
gene_name although this column is not present in the transcript database
table. By default the methods return also all columns that are used by any of
the filters submitted with the filter argument (thus, because a GenenameFilter
was used, the column gene_name is also returned). Setting
returnFilterColumns(edb) <- FALSE disables this option and only the columns
specified by the columns parameter are retrieved.
Instead of passing a filter object to the method it is also possible to provide
a filter expression written as a formula. The formula has to be written in the
form ~ <field> <condition> <value> with <field> being the field (database
column) in the database, <condition> the condition for the filter object and
<value> its value. Use the supportedFilter method to get the field names
corresponding to each filter class.
## Use a filter expression to perform the filtering.
transcripts(edb, filter = ~ genename == "ZBTB16")## GRanges object with 9 ranges and 7 metadata columns:
## seqnames ranges strand | tx_id
## <Rle> <IRanges> <Rle> | <character>
## ENST00000335953 11 114059593-114250676 + | ENST00000335953
## ENST00000541602 11 114059725-114189764 + | ENST00000541602
## ENST00000544220 11 114059737-114063646 + | ENST00000544220
## ENST00000535700 11 114060257-114063744 + | ENST00000535700
## ENST00000392996 11 114060507-114250652 + | ENST00000392996
## ENST00000539918 11 114064412-114247344 + | ENST00000539918
## ENST00000545851 11 114180766-114247296 + | ENST00000545851
## ENST00000535379 11 114237207-114250557 + | ENST00000535379
## ENST00000535509 11 114246790-114250476 + | ENST00000535509
## tx_biotype tx_cds_seq_start tx_cds_seq_end
## <character> <integer> <integer>
## ENST00000335953 protein_coding 114063301 114250555
## ENST00000541602 retained_intron <NA> <NA>
## ENST00000544220 protein_coding 114063301 114063646
## ENST00000535700 protein_coding 114063301 114063744
## ENST00000392996 protein_coding 114063301 114250555
## ENST00000539918 nonsense_mediated_decay 114064412 114121827
## ENST00000545851 processed_transcript <NA> <NA>
## ENST00000535379 processed_transcript <NA> <NA>
## ENST00000535509 retained_intron <NA> <NA>
## gene_id tx_name gene_name
## <character> <character> <character>
## ENST00000335953 ENSG00000109906 ENST00000335953 ZBTB16
## ENST00000541602 ENSG00000109906 ENST00000541602 ZBTB16
## ENST00000544220 ENSG00000109906 ENST00000544220 ZBTB16
## ENST00000535700 ENSG00000109906 ENST00000535700 ZBTB16
## ENST00000392996 ENSG00000109906 ENST00000392996 ZBTB16
## ENST00000539918 ENSG00000109906 ENST00000539918 ZBTB16
## ENST00000545851 ENSG00000109906 ENST00000545851 ZBTB16
## ENST00000535379 ENSG00000109906 ENST00000535379 ZBTB16
## ENST00000535509 ENSG00000109906 ENST00000535509 ZBTB16
## -------
## seqinfo: 1 sequence from GRCh38 genomeFilter expression have to be written as a formula (i.e. starting with a ~) in
the form column name followed by the logical condition.
Alternatively, EnsDb objects can be filtered directly using the filter
function. In the example below we use the filter function to filter the EnsDb
object and pass that filtered database to the transcripts method using the %>%
from the magrittr package.
library(magrittr)
edb %>% filter(~ symbol == "BCL2" & tx_biotype != "protein_coding") %>%
transcripts## GRanges object with 1 range and 6 metadata columns:
## seqnames ranges strand | tx_id
## <Rle> <IRanges> <Rle> | <character>
## ENST00000590515 18 63128212-63161869 - | ENST00000590515
## tx_biotype tx_cds_seq_start tx_cds_seq_end
## <character> <integer> <integer>
## ENST00000590515 processed_transcript <NA> <NA>
## gene_id tx_name
## <character> <character>
## ENST00000590515 ENSG00000171791 ENST00000590515
## -------
## seqinfo: 1 sequence from GRCh38 genomeAdding a filter to an EnsDb enables this filter (globally) on all subsequent
queries on that object. We could thus filter an EnsDb to (virtually) contain
only features encoded on chromosome Y.
edb_y <- addFilter(edb, SeqNameFilter("Y"))
## All subsequent filters on that EnsDb will only work on features encoded on
## chromosome Y
genes(edb_y)## GRanges object with 523 ranges and 6 metadata columns:
## seqnames ranges strand | gene_id
## <Rle> <IRanges> <Rle> | <character>
## ENSG00000251841 Y 2784749-2784853 + | ENSG00000251841
## ENSG00000184895 Y 2786855-2787699 - | ENSG00000184895
## ENSG00000237659 Y 2789827-2790328 + | ENSG00000237659
## ENSG00000232195 Y 2827982-2828218 + | ENSG00000232195
## ENSG00000129824 Y 2841486-2932000 + | ENSG00000129824
## ... ... ... ... . ...
## ENSG00000224240 Y 26549425-26549743 + | ENSG00000224240
## ENSG00000227629 Y 26586642-26591601 - | ENSG00000227629
## ENSG00000237917 Y 26594851-26634652 - | ENSG00000237917
## ENSG00000231514 Y 26626520-26627159 - | ENSG00000231514
## ENSG00000235857 Y 56855244-56855488 + | ENSG00000235857
## gene_name gene_biotype seq_coord_system
## <character> <character> <character>
## ENSG00000251841 RNU6-1334P snRNA chromosome
## ENSG00000184895 SRY protein_coding chromosome
## ENSG00000237659 RNASEH2CP1 processed_pseudogene chromosome
## ENSG00000232195 TOMM22P2 processed_pseudogene chromosome
## ENSG00000129824 RPS4Y1 protein_coding chromosome
## ... ... ... ...
## ENSG00000224240 CYCSP49 processed_pseudogene chromosome
## ENSG00000227629 SLC25A15P1 unprocessed_pseudogene chromosome
## ENSG00000237917 PARP4P1 unprocessed_pseudogene chromosome
## ENSG00000231514 FAM58CP processed_pseudogene chromosome
## ENSG00000235857 CTBP2P1 processed_pseudogene chromosome
## symbol entrezid
## <character> <list>
## ENSG00000251841 RNU6-1334P NA
## ENSG00000184895 SRY 6736
## ENSG00000237659 RNASEH2CP1 NA
## ENSG00000232195 TOMM22P2 NA
## ENSG00000129824 RPS4Y1 6192
## ... ... ...
## ENSG00000224240 CYCSP49 NA
## ENSG00000227629 SLC25A15P1 NA
## ENSG00000237917 PARP4P1 NA
## ENSG00000231514 FAM58CP NA
## ENSG00000235857 CTBP2P1 NA
## -------
## seqinfo: 1 sequence from GRCh38 genome## Get all lincRNAs on chromosome Y
genes(edb_y, filter = ~ gene_biotype == "lincRNA")## GRanges object with 52 ranges and 6 metadata columns:
## seqnames ranges strand | gene_id
## <Rle> <IRanges> <Rle> | <character>
## ENSG00000278847 Y 2934406-2934771 - | ENSG00000278847
## ENSG00000231535 Y 3002912-3102272 + | ENSG00000231535
## ENSG00000229308 Y 4036497-4100320 + | ENSG00000229308
## ENSG00000277930 Y 4993858-4999650 - | ENSG00000277930
## ENSG00000237069 Y 6242446-6243610 - | ENSG00000237069
## ... ... ... ... . ...
## ENSG00000228296 Y 25063083-25099892 - | ENSG00000228296
## ENSG00000223641 Y 25183643-25184773 - | ENSG00000223641
## ENSG00000228786 Y 25378300-25394719 - | ENSG00000228786
## ENSG00000240450 Y 25482908-25486705 + | ENSG00000240450
## ENSG00000231141 Y 25728490-25733388 + | ENSG00000231141
## gene_name gene_biotype seq_coord_system symbol
## <character> <character> <character> <character>
## ENSG00000278847 RP11-414C23.1 lincRNA chromosome RP11-414C23.1
## ENSG00000231535 LINC00278 lincRNA chromosome LINC00278
## ENSG00000229308 AC010084.1 lincRNA chromosome AC010084.1
## ENSG00000277930 RP11-122L9.1 lincRNA chromosome RP11-122L9.1
## ENSG00000237069 TTTY23B lincRNA chromosome TTTY23B
## ... ... ... ... ...
## ENSG00000228296 TTTY4C lincRNA chromosome TTTY4C
## ENSG00000223641 TTTY17C lincRNA chromosome TTTY17C
## ENSG00000228786 LINC00266-4P lincRNA chromosome LINC00266-4P
## ENSG00000240450 CSPG4P1Y lincRNA chromosome CSPG4P1Y
## ENSG00000231141 TTTY3 lincRNA chromosome TTTY3
## entrezid
## <list>
## ENSG00000278847 NA
## ENSG00000231535 100873962
## ENSG00000229308 NA
## ENSG00000277930 NA
## ENSG00000237069 c(100101121, 252955)
## ... ...
## ENSG00000228296 c(474150, 474149, 114761)
## ENSG00000223641 c(474152, 474151, 252949)
## ENSG00000228786 NA
## ENSG00000240450 114758
## ENSG00000231141 c(474148, 114760)
## -------
## seqinfo: 1 sequence from GRCh38 genomeTo get an overview of database tables and available columns the function
listTables can be used. The method listColumns on the other hand lists columns
for the specified database table.
## list all database tables along with their columns
listTables(edb)## $gene
## [1] "gene_id" "gene_name" "gene_biotype"
## [4] "gene_seq_start" "gene_seq_end" "seq_name"
## [7] "seq_strand" "seq_coord_system" "symbol"
##
## $tx
## [1] "tx_id" "tx_biotype" "tx_seq_start"
## [4] "tx_seq_end" "tx_cds_seq_start" "tx_cds_seq_end"
## [7] "gene_id" "tx_name"
##
## $tx2exon
## [1] "tx_id" "exon_id" "exon_idx"
##
## $exon
## [1] "exon_id" "exon_seq_start" "exon_seq_end"
##
## $chromosome
## [1] "seq_name" "seq_length" "is_circular"
##
## $protein
## [1] "tx_id" "protein_id" "protein_sequence"
##
## $uniprot
## [1] "protein_id" "uniprot_id" "uniprot_db"
## [4] "uniprot_mapping_type"
##
## $protein_domain
## [1] "protein_id" "protein_domain_id" "protein_domain_source"
## [4] "interpro_accession" "prot_dom_start" "prot_dom_end"
##
## $entrezgene
## [1] "gene_id" "entrezid"
##
## $metadata
## [1] "name" "value"## list columns from a specific table
listColumns(edb, "tx")## [1] "tx_id" "tx_biotype" "tx_seq_start"
## [4] "tx_seq_end" "tx_cds_seq_start" "tx_cds_seq_end"
## [7] "gene_id" "tx_name"Thus, we could retrieve all transcripts of the biotype nonsense_mediated_decay
(which, according to the definitions by Ensembl are transcribed, but most likely
not translated in a protein, but rather degraded after transcription) along with
the name of the gene for each transcript. Note that we are changing here the
return.type to DataFrame, so the method will return a DataFrame with the
results instead of the default GRanges.
Tx <- transcripts(edb,
columns = c(listColumns(edb , "tx"), "gene_name"),
filter = TxBiotypeFilter("nonsense_mediated_decay"),
return.type = "DataFrame")
nrow(Tx)## [1] 14423Tx## DataFrame with 14423 rows and 9 columns
## tx_id tx_biotype tx_seq_start tx_seq_end
## <character> <character> <integer> <integer>
## 1 ENST00000567466 nonsense_mediated_decay 47578 49521
## 2 ENST00000397876 nonsense_mediated_decay 53887 57372
## 3 ENST00000428730 nonsense_mediated_decay 58062 65039
## 4 ENST00000417043 nonsense_mediated_decay 62973 65037
## 5 ENST00000622194 nonsense_mediated_decay 85386 138349
## ... ... ... ... ...
## 14419 ENST00000496411 nonsense_mediated_decay 248855728 248859018
## 14420 ENST00000483223 nonsense_mediated_decay 248856515 248858529
## 14421 ENST00000533647 nonsense_mediated_decay 248857273 248858324
## 14422 ENST00000528141 nonsense_mediated_decay 248857391 248859085
## 14423 ENST00000530986 nonsense_mediated_decay 248857469 248859085
## tx_cds_seq_start tx_cds_seq_end gene_id tx_name
## <integer> <integer> <character> <character>
## 1 48546 48893 ENSG00000261456 ENST00000567466
## 2 54017 56360 ENSG00000161981 ENST00000397876
## 3 62884 65015 ENSG00000007384 ENST00000428730
## 4 63904 65015 ENSG00000007384 ENST00000417043
## 5 117330 138267 ENSG00000103148 ENST00000622194
## ... ... ... ... ...
## 14419 248857954 248858309 ENSG00000171163 ENST00000496411
## 14420 248857954 248858309 ENSG00000171163 ENST00000483223
## 14421 248857954 248858309 ENSG00000171163 ENST00000533647
## 14422 248858004 248858309 ENSG00000171163 ENST00000528141
## 14423 248858004 248858309 ENSG00000171163 ENST00000530986
## gene_name
## <character>
## 1 TUBB8
## 2 SNRNP25
## 3 RHBDF1
## 4 RHBDF1
## 5 NPRL3
## ... ...
## 14419 ZNF692
## 14420 ZNF692
## 14421 ZNF692
## 14422 ZNF692
## 14423 ZNF692For protein coding transcripts, we can also specifically extract their coding region. In the example below we extract the CDS for all transcripts encoded on chromosome Y.
yCds <- cdsBy(edb, filter = SeqNameFilter("Y"))
yCds## GRangesList object of length 151:
## $ENST00000155093
## GRanges object with 7 ranges and 3 metadata columns:
## seqnames ranges strand | seq_name exon_id
## <Rle> <IRanges> <Rle> | <character> <character>
## [1] Y 2953937-2953997 + | Y ENSE00002223884
## [2] Y 2961074-2961646 + | Y ENSE00003645989
## [3] Y 2975095-2975244 + | Y ENSE00003764421
## [4] Y 2975511-2975654 + | Y ENSE00003768468
## [5] Y 2976670-2976822 + | Y ENSE00003766362
## [6] Y 2977940-2978080 + | Y ENSE00003766086
## [7] Y 2978810-2979993 + | Y ENSE00001368923
## exon_rank
## <integer>
## [1] 2
## [2] 3
## [3] 4
## [4] 5
## [5] 6
## [6] 7
## [7] 8
##
## $ENST00000215473
## GRanges object with 2 ranges and 3 metadata columns:
## seqnames ranges strand | seq_name exon_id exon_rank
## [1] Y 5056824-5057459 + | Y ENSE00001436852 1
## [2] Y 5098215-5100740 + | Y ENSE00003741448 2
##
## $ENST00000215479
## GRanges object with 5 ranges and 3 metadata columns:
## seqnames ranges strand | seq_name exon_id exon_rank
## [1] Y 6872555-6872608 - | Y ENSE00001671586 2
## [2] Y 6870006-6870053 - | Y ENSE00001645681 3
## [3] Y 6868732-6868776 - | Y ENSE00000652250 4
## [4] Y 6868037-6868462 - | Y ENSE00001667251 5
## [5] Y 6866073-6866078 - | Y ENSE00001494454 6
##
## ...
## <148 more elements>
## -------
## seqinfo: 1 sequence from GRCh38 genomeUsing a GRangesFilter we can retrieve all features from the database that are
either within or overlapping the specified genomic region. In the example
below we query all genes that are partially overlapping with a small region on
chromosome 11. The filter restricts to all genes for which either an exon or an
intron is partially overlapping with the region.
## Define the filter
grf <- GRangesFilter(GRanges("11", ranges = IRanges(114129278, 114129328),
strand = "+"), type = "any")
## Query genes:
gn <- genes(edb, filter = grf)
gn## GRanges object with 1 range and 6 metadata columns:
## seqnames ranges strand | gene_id
## <Rle> <IRanges> <Rle> | <character>
## ENSG00000109906 11 114059593-114250676 + | ENSG00000109906
## gene_name gene_biotype seq_coord_system symbol
## <character> <character> <character> <character>
## ENSG00000109906 ZBTB16 protein_coding chromosome ZBTB16
## entrezid
## <list>
## ENSG00000109906 7704
## -------
## seqinfo: 1 sequence from GRCh38 genome## Next we retrieve all transcripts for that gene so that we can plot them.
txs <- transcripts(edb, filter = GenenameFilter(gn$gene_name))
As we can see, 4 transcripts of the gene ZBTB16 are also overlapping the
region. Below we fetch these 4 transcripts. Note, that a call to exons will
not return any features from the database, as no exon is overlapping with the
region.
transcripts(edb, filter = grf)## GRanges object with 4 ranges and 6 metadata columns:
## seqnames ranges strand | tx_id
## <Rle> <IRanges> <Rle> | <character>
## ENST00000335953 11 114059593-114250676 + | ENST00000335953
## ENST00000541602 11 114059725-114189764 + | ENST00000541602
## ENST00000392996 11 114060507-114250652 + | ENST00000392996
## ENST00000539918 11 114064412-114247344 + | ENST00000539918
## tx_biotype tx_cds_seq_start tx_cds_seq_end
## <character> <integer> <integer>
## ENST00000335953 protein_coding 114063301 114250555
## ENST00000541602 retained_intron <NA> <NA>
## ENST00000392996 protein_coding 114063301 114250555
## ENST00000539918 nonsense_mediated_decay 114064412 114121827
## gene_id tx_name
## <character> <character>
## ENST00000335953 ENSG00000109906 ENST00000335953
## ENST00000541602 ENSG00000109906 ENST00000541602
## ENST00000392996 ENSG00000109906 ENST00000392996
## ENST00000539918 ENSG00000109906 ENST00000539918
## -------
## seqinfo: 1 sequence from GRCh38 genomeThe GRangesFilter supports also GRanges defining multiple regions and a
query will return all features overlapping any of these regions. Besides using
the GRangesFilter it is also possible to search for transcripts or exons
overlapping genomic regions using the exonsByOverlaps or
transcriptsByOverlaps known from the GenomicFeatures package. Note that the
implementation of these methods for EnsDb objects supports also to use filters
to further fine-tune the query.
The functions listGenebiotypes and listTxbiotypes can be used to get an overview
of allowed/available gene and transcript biotype
## Get all gene biotypes from the database. The GeneBiotypeFilter
## allows to filter on these values.
listGenebiotypes(edb)## [1] "protein_coding"
## [2] "unitary_pseudogene"
## [3] "unprocessed_pseudogene"
## [4] "processed_pseudogene"
## [5] "processed_transcript"
## [6] "transcribed_unprocessed_pseudogene"
## [7] "antisense"
## [8] "transcribed_unitary_pseudogene"
## [9] "polymorphic_pseudogene"
## [10] "lincRNA"
## [11] "sense_intronic"
## [12] "transcribed_processed_pseudogene"
## [13] "sense_overlapping"
## [14] "IG_V_pseudogene"
## [15] "pseudogene"
## [16] "TR_V_gene"
## [17] "3prime_overlapping_ncRNA"
## [18] "IG_V_gene"
## [19] "bidirectional_promoter_lncRNA"
## [20] "snRNA"
## [21] "miRNA"
## [22] "misc_RNA"
## [23] "snoRNA"
## [24] "rRNA"
## [25] "Mt_tRNA"
## [26] "Mt_rRNA"
## [27] "IG_C_gene"
## [28] "IG_J_gene"
## [29] "TR_J_gene"
## [30] "TR_C_gene"
## [31] "TR_V_pseudogene"
## [32] "TR_J_pseudogene"
## [33] "IG_D_gene"
## [34] "ribozyme"
## [35] "IG_C_pseudogene"
## [36] "TR_D_gene"
## [37] "TEC"
## [38] "IG_J_pseudogene"
## [39] "scRNA"
## [40] "scaRNA"
## [41] "vaultRNA"
## [42] "sRNA"
## [43] "macro_lncRNA"
## [44] "non_coding"
## [45] "IG_pseudogene"
## [46] "LRG_gene"## Get all transcript biotypes from the database.
listTxbiotypes(edb)## [1] "protein_coding"
## [2] "processed_transcript"
## [3] "nonsense_mediated_decay"
## [4] "retained_intron"
## [5] "unitary_pseudogene"
## [6] "TEC"
## [7] "miRNA"
## [8] "misc_RNA"
## [9] "non_stop_decay"
## [10] "unprocessed_pseudogene"
## [11] "processed_pseudogene"
## [12] "transcribed_unprocessed_pseudogene"
## [13] "lincRNA"
## [14] "antisense"
## [15] "transcribed_unitary_pseudogene"
## [16] "polymorphic_pseudogene"
## [17] "sense_intronic"
## [18] "transcribed_processed_pseudogene"
## [19] "sense_overlapping"
## [20] "IG_V_pseudogene"
## [21] "pseudogene"
## [22] "TR_V_gene"
## [23] "3prime_overlapping_ncRNA"
## [24] "IG_V_gene"
## [25] "bidirectional_promoter_lncRNA"
## [26] "snRNA"
## [27] "snoRNA"
## [28] "rRNA"
## [29] "Mt_tRNA"
## [30] "Mt_rRNA"
## [31] "IG_C_gene"
## [32] "IG_J_gene"
## [33] "TR_J_gene"
## [34] "TR_C_gene"
## [35] "TR_V_pseudogene"
## [36] "TR_J_pseudogene"
## [37] "IG_D_gene"
## [38] "ribozyme"
## [39] "IG_C_pseudogene"
## [40] "TR_D_gene"
## [41] "IG_J_pseudogene"
## [42] "scRNA"
## [43] "scaRNA"
## [44] "vaultRNA"
## [45] "sRNA"
## [46] "macro_lncRNA"
## [47] "non_coding"
## [48] "IG_pseudogene"
## [49] "LRG_gene"Data can be fetched in an analogous way using the exons and genes
methods. In the example below we retrieve gene_name, entrezid and the
gene_biotype of all genes in the database which names start with “BCL2”.
## We're going to fetch all genes which names start with BCL.
BCLs <- genes(edb,
columns = c("gene_name", "entrezid", "gene_biotype"),
filter = GenenameFilter("BCL", condition = "startsWith"),
return.type = "DataFrame")
nrow(BCLs)## [1] 30BCLs## DataFrame with 30 rows and 4 columns
## gene_name entrezid gene_biotype gene_id
## <character> <list> <character> <character>
## 1 BCL10 8915 protein_coding ENSG00000142867
## 2 BCL11A 53335 protein_coding ENSG00000119866
## 3 BCL11B 64919 protein_coding ENSG00000127152
## 4 BCL2 596 protein_coding ENSG00000171791
## 5 BCL2A1 597 protein_coding ENSG00000140379
## ... ... ... ... ...
## 26 BCL9L 283149 protein_coding ENSG00000186174
## 27 BCL9P1 NA processed_pseudogene ENSG00000249238
## 28 BCLAF1 9774 protein_coding ENSG00000029363
## 29 BCLAF1P1 NA processed_pseudogene ENSG00000248966
## 30 BCLAF1P2 NA processed_pseudogene ENSG00000279800Sometimes it might be useful to know the length of genes or transcripts
(i.e. the total sum of nucleotides covered by their exons). Below we calculate
the mean length of transcripts from protein coding genes on chromosomes X and Y
as well as the average length of snoRNA, snRNA and rRNA transcripts encoded on
these chromosomes. For the first query we combine two AnnotationFilter objects
using an AnnotationFilterList object, in the second we define the query using a
filter expression.
## determine the average length of snRNA, snoRNA and rRNA genes encoded on
## chromosomes X and Y.
mean(lengthOf(edb, of = "tx",
filter = AnnotationFilterList(
GeneBiotypeFilter(c("snRNA", "snoRNA", "rRNA")),
SeqNameFilter(c("X", "Y")))))## [1] 118.2458## determine the average length of protein coding genes encoded on the same
## chromosomes.
mean(lengthOf(edb, of = "tx",
filter = ~ gene_biotype == "protein_coding" &
seq_name %in% c("X", "Y")))## [1] 1943.554Not unexpectedly, transcripts of protein coding genes are longer than those of snRNA, snoRNA or rRNA genes.
At last we extract the first two exons of each transcript model from the database.
## Extract all exons 1 and (if present) 2 for all genes encoded on the
## Y chromosome
exons(edb, columns = c("tx_id", "exon_idx"),
filter = list(SeqNameFilter("Y"),
ExonRankFilter(3, condition = "<")))## GRanges object with 1294 ranges and 3 metadata columns:
## seqnames ranges strand | tx_id
## <Rle> <IRanges> <Rle> | <character>
## ENSE00002088309 Y 2784749-2784853 + | ENST00000516032
## ENSE00001494622 Y 2786855-2787699 - | ENST00000383070
## ENSE00001772499 Y 2789827-2790328 + | ENST00000454281
## ENSE00001614266 Y 2827982-2828218 + | ENST00000430735
## ENSE00002490412 Y 2841486-2841627 + | ENST00000250784
## ... ... ... ... . ...
## ENSE00001632993 Y 26591548-26591601 - | ENST00000456738
## ENSE00001616687 Y 26626520-26627159 - | ENST00000435741
## ENSE00001638296 Y 26633345-26633431 - | ENST00000435945
## ENSE00001797328 Y 26634523-26634652 - | ENST00000435945
## ENSE00001794473 Y 56855244-56855488 + | ENST00000431853
## exon_idx exon_id
## <integer> <character>
## ENSE00002088309 1 ENSE00002088309
## ENSE00001494622 1 ENSE00001494622
## ENSE00001772499 1 ENSE00001772499
## ENSE00001614266 1 ENSE00001614266
## ENSE00002490412 1 ENSE00002490412
## ... ... ...
## ENSE00001632993 1 ENSE00001632993
## ENSE00001616687 1 ENSE00001616687
## ENSE00001638296 2 ENSE00001638296
## ENSE00001797328 1 ENSE00001797328
## ENSE00001794473 1 ENSE00001794473
## -------
## seqinfo: 1 sequence from GRCh38 genome3 Extracting gene/transcript/exon models for RNASeq feature counting
For the feature counting step of an RNAseq experiment, the gene or transcript
models (defined by the chromosomal start and end positions of their exons) have
to be known. To extract these from an Ensembl based annotation package, the
exonsBy, genesBy and transcriptsBy methods can be used in an analogous way as in
TxDb packages generated by the GenomicFeatures package. However, the
transcriptsBy method does not, in contrast to the method in the GenomicFeatures
package, allow to return transcripts by “cds”. While the annotation packages
built by the ensembldb contain the chromosomal start and end coordinates of
the coding region (for protein coding genes) they do not assign an ID to each
CDS.
A simple use case is to retrieve all genes encoded on chromosomes X and Y from the database.
TxByGns <- transcriptsBy(edb, by = "gene", filter = SeqNameFilter(c("X", "Y")))
TxByGns## GRangesList object of length 2922:
## $ENSG00000000003
## GRanges object with 5 ranges and 6 metadata columns:
## seqnames ranges strand | tx_id
## <Rle> <IRanges> <Rle> | <character>
## [1] X 100633442-100639991 - | ENST00000494424
## [2] X 100627109-100637104 - | ENST00000612152
## [3] X 100632063-100637104 - | ENST00000614008
## [4] X 100628670-100636806 - | ENST00000373020
## [5] X 100632541-100636689 - | ENST00000496771
## tx_biotype tx_cds_seq_start tx_cds_seq_end gene_id
## <character> <integer> <integer> <character>
## [1] processed_transcript <NA> <NA> ENSG00000000003
## [2] protein_coding 100630798 100635569 ENSG00000000003
## [3] protein_coding 100632063 100635569 ENSG00000000003
## [4] protein_coding 100630798 100636694 ENSG00000000003
## [5] processed_transcript <NA> <NA> ENSG00000000003
## tx_name
## <character>
## [1] ENST00000494424
## [2] ENST00000612152
## [3] ENST00000614008
## [4] ENST00000373020
## [5] ENST00000496771
##
## $ENSG00000000005
## GRanges object with 2 ranges and 6 metadata columns:
## seqnames ranges strand | tx_id
## [1] X 100584802-100599885 + | ENST00000373031
## [2] X 100593624-100597531 + | ENST00000485971
## tx_biotype tx_cds_seq_start tx_cds_seq_end gene_id
## [1] protein_coding 100585019 100599717 ENSG00000000005
## [2] processed_transcript <NA> <NA> ENSG00000000005
## tx_name
## [1] ENST00000373031
## [2] ENST00000485971
##
## $ENSG00000001497
## GRanges object with 5 ranges and 6 metadata columns:
## seqnames ranges strand | tx_id
## [1] X 65512583-65534775 - | ENST00000484069
## [2] X 65512582-65534756 - | ENST00000374811
## [3] X 65512583-65534756 - | ENST00000374804
## [4] X 65512582-65534754 - | ENST00000374807
## [5] X 65520429-65523617 - | ENST00000469091
## tx_biotype tx_cds_seq_start tx_cds_seq_end
## [1] nonsense_mediated_decay 65525021 65534715
## [2] protein_coding 65512775 65534715
## [3] protein_coding 65512775 65534715
## [4] protein_coding 65512775 65534715
## [5] protein_coding 65520655 65523617
## gene_id tx_name
## [1] ENSG00000001497 ENST00000484069
## [2] ENSG00000001497 ENST00000374811
## [3] ENSG00000001497 ENST00000374804
## [4] ENSG00000001497 ENST00000374807
## [5] ENSG00000001497 ENST00000469091
##
## ...
## <2919 more elements>
## -------
## seqinfo: 2 sequences from GRCh38 genomeSince Ensembl contains also definitions of genes that are on chromosome variants (supercontigs), it is advisable to specify the chromosome names for which the gene models should be returned.
In a real use case, we might thus want to retrieve all genes encoded on the
standard chromosomes. In addition it is advisable to use a GeneIdFilter to
restrict to Ensembl genes only, as also LRG (Locus Reference Genomic)
genes2 are defined in the database, which are partially redundant with
Ensembl genes.
## will just get exons for all genes on chromosomes 1 to 22, X and Y.
## Note: want to get rid of the "LRG" genes!!!
EnsGenes <- exonsBy(edb, by = "gene", filter = AnnotationFilterList(
SeqNameFilter(c(1:22, "X", "Y")),
GeneIdFilter("ENSG", "startsWith")))The code above returns a GRangesList that can be used directly as an input for
the summarizeOverlaps function from the GenomicAlignments package 3.
Alternatively, the above GRangesList can be transformed to a data.frame in
SAF format that can be used as an input to the featureCounts function of the
Rsubread package 4.
## Transforming the GRangesList into a data.frame in SAF format
EnsGenes.SAF <- toSAF(EnsGenes)Note that the ID by which the GRangesList is split is used in the SAF
formatted data.frame as the GeneID. In the example below this would be the
Ensembl gene IDs, while the start, end coordinates (along with the strand and
chromosomes) are those of the the exons.
In addition, the disjointExons function (similar to the one defined in
GenomicFeatures) can be used to generate a GRanges of non-overlapping exon
parts which can be used in the DEXSeq package.
## Create a GRanges of non-overlapping exon parts.
DJE <- disjointExons(edb, filter = AnnotationFilterList(
SeqNameFilter(c(1:22, "X", "Y")),
GeneIdFilter("ENSG%", "startsWith")))4 Retrieving sequences for gene/transcript/exon models
The methods to retrieve exons, transcripts and genes (i.e. exons, transcripts
and genes) return by default GRanges objects that can be used to retrieve
sequences using the getSeq method e.g. from BSgenome packages. The basic
workflow is thus identical to the one for TxDb packages, however, it is not
straight forward to identify the BSgenome package with the matching genomic
sequence. Most BSgenome packages are named according to the genome build
identifier used in UCSC which does not (always) match the genome build name used
by Ensembl. Using the Ensembl version provided by the EnsDb, the correct genomic
sequence can however be retrieved easily from the AnnotationHub using the
getGenomeFaFile. If no Fasta file matching the Ensembl version is available, the
function tries to identify a Fasta file with the correct genome build from the
closest Ensembl release and returns that instead.
In the code block below we retrieve first the FaFile with the genomic DNA
sequence, extract the genomic start and end coordinates for all genes defined in
the package, subset to genes encoded on sequences available in the FaFile and
extract all of their sequences. Note: these sequences represent the sequence
between the chromosomal start and end coordinates of the gene.
library(EnsDb.Hsapiens.v86)
library(Rsamtools)
edb <- EnsDb.Hsapiens.v86
## Get the FaFile with the genomic sequence matching the Ensembl version
## using the AnnotationHub package.
Dna <- getGenomeFaFile(edb)
## Get start/end coordinates of all genes.
genes <- genes(edb)
## Subset to all genes that are encoded on chromosomes for which
## we do have DNA sequence available.
genes <- genes[seqnames(genes) %in% seqnames(seqinfo(Dna))]
## Get the gene sequences, i.e. the sequence including the sequence of
## all of the gene's exons and introns.
geneSeqs <- getSeq(Dna, genes)To retrieve the (exonic) sequence of transcripts (i.e. without introns) we can
use directly the extractTranscriptSeqs method defined in the GenomicFeatures on
the EnsDb object, eventually using a filter to restrict the query.
## get all exons of all transcripts encoded on chromosome Y
yTx <- exonsBy(edb, filter = SeqNameFilter("Y"))
## Retrieve the sequences for these transcripts from the FaFile.
library(GenomicFeatures)
yTxSeqs <- extractTranscriptSeqs(Dna, yTx)
yTxSeqs
## Extract the sequences of all transcripts encoded on chromosome Y.
yTx <- extractTranscriptSeqs(Dna, edb, filter = SeqNameFilter("Y"))
## Along these lines, we could use the method also to retrieve the coding sequence
## of all transcripts on the Y chromosome.
cdsY <- cdsBy(edb, filter = SeqNameFilter("Y"))
extractTranscriptSeqs(Dna, cdsY)Next we retrieve transcript sequences from genes encoded on chromosome Y using
the BSGenome package for the human genome. Ensembl version 86 based on
the GRCh38 genome build and we thus load the corresponding BSGenome package.
library(BSgenome.Hsapiens.NCBI.GRCh38)
bsg <- BSgenome.Hsapiens.NCBI.GRCh38
## Get the genome version
unique(genome(bsg))## [1] "GRCh38"unique(genome(edb))## [1] "GRCh38"## Extract the full transcript sequences.
yTxSeqs <- extractTranscriptSeqs(bsg, exonsBy(edb, "tx",
filter = SeqNameFilter("Y")))
yTxSeqs## A DNAStringSet instance of length 740
## width seq names
## [1] 5239 GCCTAGTGCGCGCGCAGTAAC...TAAATGTTTACTTGTATATG ENST00000155093
## [2] 4595 CTGGTGGTCCAGTACCTCCAA...AGCCCTTCAGAAGACATTCT ENST00000215473
## [3] 802 AGAGGACCAAGCCTCCCTGTG...ATAAAATGTTTTAAAAATCA ENST00000215479
## [4] 910 TGTCTGTCAGAGCTGTCAGCC...AACACTGGTATATTTCTGTT ENST00000250776
## [5] 1305 TTCCAGGATATGAACTCTACA...AATCCTGTGGCTGTAGGAAA ENST00000250784
## ... ... ...
## [736] 792 ATGGCCCGGGGCCCCAAGAAG...CCAAACAGAGCAGTGGCTAA ENST00000629237
## [737] 344 GGTTGCCACTTCAAGGGACTA...GGCTCTTCTGGCAGTTTTTT ENST00000631331
## [738] 933 CTCTCCCAGCTTCTACCCACA...ATACTATAAAAATGCTTTAA ENST00000634531
## [739] 1832 ATGTCTGCTGCAAATCCTGAG...TATTTAAATCTGTTGGATCC ENST00000634662
## [740] 890 CTCTCCCAGCTTCTACCCACA...ATACTATAAAAATGCTTTAA ENST00000635343## Extract just the CDS
Test <- cdsBy(edb, "tx", filter = SeqNameFilter("Y"))
yTxCds <- extractTranscriptSeqs(bsg, cdsBy(edb, "tx",
filter = SeqNameFilter("Y")))
yTxCds## A DNAStringSet instance of length 151
## width seq names
## [1] 2406 ATGGATGAAGATGAATTTGAA...AAGAAGTTGGTCTGCCCTAA ENST00000155093
## [2] 3162 ATGTTTAGGGTTGGCTTCTTA...TTTCTAACACAACTTTCTAA ENST00000215473
## [3] 579 ATGGGGACCTGGATTTTGTTT...AGCAGGAGGAAGTGGATTAA ENST00000215479
## [4] 792 ATGGCCCGGGGCCCCAAGAAG...CCAAACAGAGCAGTGGCTAA ENST00000250784
## [5] 378 ATGAGTCCAAAGCCGAGAGCC...CTACTCCCCTATCTCCCTGA ENST00000250823
## ... ... ...
## [147] 387 ATGCAAAGCCAGAGAGGTCTC...CACTCTGTGTCCCAAAATGA ENST00000624507
## [148] 78 ATGAGAGCCAAGTGGAGGAAG...TGAGGCAGAAGTCCAAGTAA ENST00000624575
## [149] 1833 ATGGATGAAGATGAATTTGAA...AAGAAGTTGGTCTGCCCTAA ENST00000625061
## [150] 792 ATGGCCCGGGGCCCCAAGAAG...CCAAACAGAGCAGTGGCTAA ENST00000629237
## [151] 1740 ATGTCTGCTGCAAATCCTGAG...TAATCCAGAGAAGAGACTGA ENST000006346625 Integrating annotations from Ensembl based EnsDb packages with UCSC based annotations
Sometimes it might be useful to combine (Ensembl based) annotations from EnsDb
packages/objects with annotations from other Bioconductor packages, that might
base on UCSC annotations. To support such an integration of annotations, the
ensembldb packages implements the seqlevelsStyle and seqlevelsStyle<- from the
GenomeInfoDb package that allow to change the style of chromosome naming. Thus,
sequence/chromosome names other than those used by Ensembl can be used in, and
are returned by, the queries to EnsDb objects as long as a mapping for them is
provided by the GenomeInfoDb package (which provides a mapping mostly between
UCSC, NCBI and Ensembl chromosome names for the main chromosomes).
In the example below we change the seqnames style to UCSC.
## Change the seqlevels style form Ensembl (default) to UCSC:
seqlevelsStyle(edb) <- "UCSC"
## Now we can use UCSC style seqnames in SeqNameFilters or GRangesFilter:
genesY <- genes(edb, filter = ~ seq_name == "chrY")
## The seqlevels of the returned GRanges are also in UCSC style
seqlevels(genesY)## [1] "chrY"Note that in most instances no mapping is available for sequences not
corresponding to the main chromosomes (i.e. contigs, patched chromosomes
etc). What is returned in cases in which no mapping is available can be
specified with the global ensembldb.seqnameNotFound option. By default (with
ensembldb.seqnameNotFound set to “ORIGINAL”), the original seqnames (i.e. the
ones from Ensembl) are returned. With ensembldb.seqnameNotFound “MISSING” each
time a seqname can not be found an error is thrown. For all other cases
(e.g. ensembldb.seqnameNotFound = NA) the value of the option is returned.
seqlevelsStyle(edb) <- "UCSC"
## Getting the default option:
getOption("ensembldb.seqnameNotFound")## [1] "ORIGINAL"## Listing all seqlevels in the database.
seqlevels(edb)[1:30]## Warning in .formatSeqnameByStyleFromQuery(x, sn, ifNotFound): More than 5
## seqnames with seqlevels style of the database (Ensembl) could not be mapped
## to the seqlevels style: UCSC!) Returning the orginal seqnames for these.## [1] "chr1" "chr10"
## [3] "chr11" "chr12"
## [5] "chr13" "chr14"
## [7] "chr15" "chr16"
## [9] "chr17" "chr18"
## [11] "chr19" "chr2"
## [13] "chr20" "chr21"
## [15] "chr22" "chr3"
## [17] "chr4" "chr5"
## [19] "chr6" "chr7"
## [21] "chr8" "chr9"
## [23] "CHR_HG107_PATCH" "CHR_HG126_PATCH"
## [25] "CHR_HG1311_PATCH" "CHR_HG1342_HG2282_PATCH"
## [27] "CHR_HG1362_PATCH" "CHR_HG142_HG150_NOVEL_TEST"
## [29] "CHR_HG151_NOVEL_TEST" "CHR_HG1651_PATCH"## Setting the option to NA, thus, for each seqname for which no mapping is available,
## NA is returned.
options(ensembldb.seqnameNotFound=NA)
seqlevels(edb)[1:30]## Warning in .formatSeqnameByStyleFromQuery(x, sn, ifNotFound): More than 5
## seqnames with seqlevels style of the database (Ensembl) could not be mapped
## to the seqlevels style: UCSC!) Returning NA for these.## [1] "chr1" "chr10" "chr11" "chr12" "chr13" "chr14" "chr15" "chr16" "chr17"
## [10] "chr18" "chr19" "chr2" "chr20" "chr21" "chr22" "chr3" "chr4" "chr5"
## [19] "chr6" "chr7" "chr8" "chr9" NA NA NA NA NA
## [28] NA NA NA## Resetting the option.
options(ensembldb.seqnameNotFound = "ORIGINAL")At last changing the seqname style to the default value "Ensembl".
seqlevelsStyle(edb) <- "Ensembl"6 Interactive annotation lookup using the shiny web app
In addition to the genes, transcripts and exons methods it is possibly to
search interactively for gene/transcript/exon annotations using the internal,
shiny based, web application. The application can be started with the
runEnsDbApp() function. The search results from this app can also be returned
to the R workspace either as a data.frame or GRanges object.
7 Plotting gene/transcript features using ensembldb and Gviz and ggbio
The Gviz package provides functions to plot genes and transcripts along with
other data on a genomic scale. Gene models can be provided either as a
data.frame, GRanges, TxDB database, can be fetched from biomart and can
also be retrieved from ensembldb.
Below we generate a GeneRegionTrack fetching all transcripts from a certain
region on chromosome Y.
Note that if we want in addition to work also with BAM files that were aligned
against DNA sequences retrieved from Ensembl or FASTA files representing genomic
DNA sequences from Ensembl we should change the ucscChromosomeNames option from
Gviz to FALSE (i.e. by calling options(ucscChromosomeNames = FALSE)). This is
not necessary if we just want to retrieve gene models from an EnsDb object, as
the ensembldb package internally checks the ucscChromosomeNames option and,
depending on that, maps Ensembl chromosome names to UCSC chromosome names.
## Loading the Gviz library
library(Gviz)
library(EnsDb.Hsapiens.v86)
edb <- EnsDb.Hsapiens.v86
## Retrieving a Gviz compatible GRanges object with all genes
## encoded on chromosome Y.
gr <- getGeneRegionTrackForGviz(edb, chromosome = "Y",
start = 20400000, end = 21400000)
## Define a genome axis track
gat <- GenomeAxisTrack()
## We have to change the ucscChromosomeNames option to FALSE to enable Gviz usage
## with non-UCSC chromosome names.
options(ucscChromosomeNames = FALSE)
plotTracks(list(gat, GeneRegionTrack(gr)))
options(ucscChromosomeNames = TRUE)Above we had to change the option ucscChromosomeNames to FALSE in order to
use it with non-UCSC chromosome names. Alternatively, we could however also
change the seqnamesStyle of the EnsDb object to UCSC. Note that we have to
use now also chromosome names in the UCSC style in the SeqNameFilter
(i.e. “chrY” instead of “Y”).
seqlevelsStyle(edb) <- "UCSC"
## Retrieving the GRanges objects with seqnames corresponding to UCSC chromosome names.
gr <- getGeneRegionTrackForGviz(edb, chromosome = "chrY",
start = 20400000, end = 21400000)## Warning in .formatSeqnameByStyleForQuery(x, sn, ifNotFound): Seqnames:
## Y could not be mapped to the seqlevels style of the database (Ensembl)!
## Returning the orginal seqnames for these.seqnames(gr)## factor-Rle of length 89 with 1 run
## Lengths: 89
## Values : chrY
## Levels(1): chrY## Define a genome axis track
gat <- GenomeAxisTrack()
plotTracks(list(gat, GeneRegionTrack(gr)))
We can also use the filters from the ensembldb package to further refine what
transcripts are fetched, like in the example below, in which we create two
different gene region tracks, one for protein coding genes and one for lincRNAs.
protCod <- getGeneRegionTrackForGviz(edb, chromosome = "chrY",
start = 20400000, end = 21400000,
filter = GeneBiotypeFilter("protein_coding"))
lincs <- getGeneRegionTrackForGviz(edb, chromosome = "chrY",
start = 20400000, end = 21400000,
filter = GeneBiotypeFilter("lincRNA"))
plotTracks(list(gat, GeneRegionTrack(protCod, name = "protein coding"),
GeneRegionTrack(lincs, name = "lincRNAs")), transcriptAnnotation = "symbol")
## At last we change the seqlevels style again to Ensembl
seqlevelsStyle <- "Ensembl"Alternatively, we can also use ggbio for plotting. For ggbio we can directly
pass the EnsDb object along with optional filters (or as in the example below a
filter expression as a formula).
library(ggbio)
## Create a plot for all transcripts of the gene SKA2
autoplot(edb, ~ genename == "SKA2")