The package taxlist
aims to implement an object class
and functions (methods) for handling taxonomic data in
R. The homonymous object class taxlist
can
be further linked to biodiversity records (e.g. for observations in
vegetation plots).
The taxlist
package is developed on the repository
GitHub (https://github.com/ropensci/taxlist)
and can be installed in your R-session using the package
devtools
:
Since this package is already available in the Comprehensive R
Archive Network (CRAN), it is also possible to install it using the
function install.packages
:
Of course, you have to load taxlist
into your
R-session.
For accessing to this vignette, use following command:
One of the main tasks of taxlist
is to structure
taxonomic information for a further linkage to biodiversity records.
This structure have to be on the one side consistent with taxonomic
issues (e.g. synonyms, hierarchies, etc.), on the other side have to be
flexible for containing different depth of information availability
(from plain species lists to hierarchical structures).
In this guide, we will work with a species list from phytosociological relevés collected at the borderline between the Democratic Republic of the Congo and Rwanda (Mullenders 1953 Vegetatio 4(2): 73–83).
The digitized data can be loaded by following command:
The data is formatted as data.frame
in
R, including the names of the species in the first
column:
## TaxonName 3094 3093 3092 3095 3096 3097 3098
## 1 Eragrostis tenuifolia + <NA> <NA> <NA> <NA> <NA> <NA>
## 2 Cyperus sublimis <NA> + <NA> <NA> <NA> <NA> <NA>
## 3 Digitaria abyssinica + 1 2 2 2 3 1
## 4 Hyparrhenia filipendula <NA> <NA> <NA> <NA> <NA> <NA> <NA>
## 5 Erigeron floribundus + 1 <NA> <NA> <NA> <NA> <NA>
## 6 Aerva lanata + 1 <NA> <NA> <NA> <NA> <NA>
As already mentioned, the first column in the cross table contains
the names of the species occurring in the observed plots. Thus, we can
use this character vector to construct a taxlist
object.
This can be achieved through the function df2taxlist()
.
## Missing column 'TaxonConceptID' in 'x'. All names will be considered as accepted names.
## object size: 9 Kb
## validation of 'taxlist' object: TRUE
##
## number of taxon usage names: 35
## number of taxon concepts: 35
## trait entries: 0
## number of trait variables: 0
## taxon views: 0
Note that the function summary
provides a quick overview
in the content of the resulting object. This function can be also
applied to a specific taxon:
## ------------------------------
## concept ID: 5
## view ID: none
## level: none
## parent: none
##
## # accepted name:
## 5 Erigeron floribundus NA
## ------------------------------
The installation of taxlist
includes the data
Easplist
, which is formatted as a taxlist
object. This data is a subset of the species list used by the database
SWEA-Dataveg (GIVD ID
AF-006):
## object size: 761.4 Kb
## validation of 'taxlist' object: TRUE
##
## number of taxon usage names: 5393
## number of taxon concepts: 3887
## trait entries: 311
## number of trait variables: 1
## taxon views: 3
##
## concepts with parents: 3698
## concepts with children: 1343
##
## hierarchical levels: form < variety < subspecies < species < complex < genus < family
## number of concepts in level form: 2
## number of concepts in level variety: 95
## number of concepts in level subspecies: 71
## number of concepts in level species: 2521
## number of concepts in level complex: 1
## number of concepts in level genus: 1011
## number of concepts in level family: 186
The common ways to access to the content of slots in S4
objects are either using the function slot(object, name)
or
the symbol @
(i.e. object@name
). Additional
functions, which are specific for taxlist
objects are
taxon_names
, taxon_relations
,
taxon_traits
and taxon_views
(see the help
documentation).
Additionally, it is possible to use the methods $
and
[
, the first for access to information in the slot
taxonTraits
, while the second can be also used for other
slots in the object.
## acropleustophyte chamaephyte climbing_plant facultative_annual
## 8 25 25 20
## obligate_annual phanerophyte pleustohelophyte reed_plant
## 114 26 8 14
## reptant_plant tussock_plant NA's
## 19 52 3576
Methods for the function subset
are also implemented in
this package. Such subsets usually apply pattern matching (for character
vectors) or logical operations and are analogous to query building in
relational databases. The subset
method can be apply to any
slot by setting the value of the argument slot
.
Papyrus <- subset(x = Easplist, subset = grepl("papyrus", TaxonName), slot = "names")
summary(Papyrus, "all")
Or the very same results:
Papyrus <- subset(x = Easplist, subset = TaxonConceptID == 206, slot = "relations")
summary(Papyrus, "all")
Similarly, you can look for a specific name.
Objects belonging to the class taxlist
can optionally
content parent-child relationships and taxonomic levels. Such
information is also included in the data Easplist
, as shown
in the summary output.
## object size: 761.4 Kb
## validation of 'taxlist' object: TRUE
##
## number of taxon usage names: 5393
## number of taxon concepts: 3887
## trait entries: 311
## number of trait variables: 1
## taxon views: 3
##
## concepts with parents: 3698
## concepts with children: 1343
##
## hierarchical levels: form < variety < subspecies < species < complex < genus < family
## number of concepts in level form: 2
## number of concepts in level variety: 95
## number of concepts in level subspecies: 71
## number of concepts in level species: 2521
## number of concepts in level complex: 1
## number of concepts in level genus: 1011
## number of concepts in level family: 186
Note that such information can get lost once subset()
has been applied, since the respective parents or children from the
original data set are not anymore in the subset. May you like to recover
parents and children, you can use the functions
get_parents()
or get_children()
,
respectively.
## ------------------------------
## concept ID: 206
## view ID: 1
## level: species
## parent: none
##
## # accepted name:
## 206 Cyperus papyrus L.
##
## # synonyms (2):
## 52612 Cyperus papyrus ssp. antiquorum (Willd.) Chiov.
## 52613 Cyperus papyrus ssp. nyassicus Chiov.
## ------------------------------
## ------------------------------
## concept ID: 206
## view ID: 1
## level: species
## parent: 54853 Cyperus L.
##
## # accepted name:
## 206 Cyperus papyrus L.
##
## # synonyms (2):
## 52612 Cyperus papyrus ssp. antiquorum (Willd.) Chiov.
## 52613 Cyperus papyrus ssp. nyassicus Chiov.
## ------------------------------
## concept ID: 54853
## view ID: 2
## level: genus
## parent: 55959 Cyperaceae Juss.
##
## # accepted name:
## 54855 Cyperus L.
## ------------------------------
## concept ID: 55959
## view ID: 3
## level: family
## parent: none
##
## # accepted name:
## 55961 Cyperaceae Juss.
## ------------------------------
Another way to represent taxonomic ranks is by using the function
indented_list()
.
## Cyperaceae Juss.
## Cyperus L.
## Cyperus papyrus L.
To illustrate the flexibility of the taxlist
objects,
the next example will handle a syntaxonomical scheme. As example it will
be used a scheme proposed by the author for aquatic and semi-aquatic
vegetation in Tanzania (Alvarez 2017 Phytocoenologia in
review). The scheme includes 10 associations classified into 4
classes: