Using the "select" interface on TxDb objects
select, columns and keys can be used together to
extract data from a TxDb object.
In the code snippets below, x is a TxDb object.
keytypes(x):
allows the user to discover which keytypes can be passed in to
select or keys and the keytype argument.
keys(x, keytype, pattern, column, fuzzy): Return keys for
the database contained in the TxDb object .
The keytype argument specifies the kind of keys that will
be returned. By default keys will return the "GENEID" keys
for the database.
If keys is used with pattern, it will pattern match
on the keytype.
But if the column argument is also provided along with the
pattern argument, then pattern will be matched
against the values in column instead.
And if keys is called with column and no
pattern argument, then it will return all keys that have
corresponding values in the column argument.
Thus, the behavior of keys all depends on how many arguments are
specified.
Use of the fuzzy argument will toggle fuzzy matching to
TRUE or FALSE. If pattern is not used, fuzzy is ignored.
columns(x):
Show which kinds of data can be returned for the
TxDb object.
select(x, keys, columns, keytype):
When all the appropriate arguments are specified select
will retrieve the matching data as a data.frame based on
parameters for selected keys and columns and
keytype arguments.
Marc Carlson
AnnotationDb-class for more descriptsion
of methods select,keytypes,keys and columns.
transcripts, transcriptsBy,
and transcriptsByOverlaps, for other ways to
extract genomic features from a TxDb object.
The TxDb class.
txdb_file <- system.file("extdata", "Biomart_Ensembl_sample.sqlite",
package="GenomicFeatures")
txdb <- loadDb(txdb_file)
txdb
## find key types
keytypes(txdb)
## list IDs that can be used to filter
head(keys(txdb, "GENEID"))
head(keys(txdb, "TXID"))
head(keys(txdb, "TXNAME"))
## list columns that can be returned by select
columns(txdb)
## call select
res <- select(txdb, head(keys(txdb, "GENEID")),
columns=c("GENEID","TXNAME"),
keytype="GENEID")
head(res)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.