datamuseum is an R package designed to improve specimen
management and improve access to legacy databases. This extends from
first steps, such as removing samples with unclear or missing labels, to
aesthetic improvements intended for publishable tables and figures.
datamuseum is categorized into four main parts, and each
will be described briefly here. Please refer to our Articles
for more in-depth workflows and breakdowns.
Geographic Data
datamuseum includes a host of functions designed for
manipulating and refining geographic data. Although the package as a
whole is intended for biological samples, functions in the
latlong_* group are universally applicable.
Management with datamuseum is done in part with rnaturalearth,
which provides a standard reference for geography and works alongside
other functions of the package to ensure mappable data is consistent and
ready for display with other packages like ggplot2.
Taxonomic Data
At its core, datamuseum is intended for use with
biological specimen data sets, with a focus on taxonomy.
datamuseum accomplishes the task of robust taxonomic review
in its taxon function group through a dual-reference system which checks
against the Global Biodiversity Information Facility’s rGBIF package,
and the Integrated Taxonomic Information System via taxize. Users have
the option to use GBIF, ITIS, or both for their taxonomic inquiries.
datamuseum is only as good as its sources, and user
input is occasionally essential. Although its ability to double-check
provides safety against misspellings or inconsistent updates, some cases
may still arise where manual changes are still necessary.
Example Data
In lieu of its limitations, datamuseum includes practice
data in the form of accessioned octopus specimen data sets refined to
Japan. Superfamily Octopodoidea was selected and refined to Japan due to
the long history of octopus fisheries and cuisine in the region, as well
as recent regional taxonomic updates.
More specifically, Octopus vulgaris (Cuvier, 1797) in East
Asia was recently re-described as the also valid Octopus
sinensis (d’Orbigny, 1834) by Gleadall
et al. (2016). datamuseum would struggle to reflect a
regional change like this since both species are still valid and
accepted by GBIF and ITIS. As a result, we recommend adding a simple
command like the one below to your datamuseum workflow for
case-specific changes.
This code, and its broader workflow with associated
taxon_* and latlong_* functions, can also be
found at Octopodoidea
in Japan.
museum_clean <- museum_clean %>%
mutate(SciName = case_when(
SciName == "Octopus vulgaris" ~ "Octopus sinensis",
TRUE ~ SciName
))Utilities
Going beyond manipulating just geographic and taxonomic data,
datamuseum also provides management options for accessioned
data sets. This includes options for removing repeat data
(deduplicating) when multiple sources are accessioned, expanding your
data set when one specimen lot contains multiple individuals, and
preparing your taxonomic data for presentation on a ggplot2 graph!