respR was designed to be an end-to-end solution for
analysing data and reporting analyses from any and all aquatic
respirometry experiments regardless of the equipment used. Therefore, it
is system-agnostic; the data need only be put into a very simple
structure - numeric time against oxygen in any common units in a
data.frame - to allow a full analysis to be conducted. In
fact the entire package, with the exception of the final conversion step
in convert_rate, treats data as unitless so non-aquatic
respirometry data or any
time-series data can be explored and analysed.
To our knowledge, every oxygen probe system allows data to be
exported in easily readable formats (e.g. .csv,
.txt, .xlsx) which contain the numeric time
and oxygen data respR requires. These files are best
imported using generic functions after which the relevant columns can be
specified when used in respR functions.
There is a wealth of information online about importing data to
R. See here, here and here for
just a few examples.
The following functions may help with importing your data:
read.csv(), read.table(),
read.delim() - Generic base R import
functions. Also import .txt files. The nrows
and skip inputs are particularly useful for specifying how
many rows to import and how many to skip at the start of the
file.
readr - A
Tidyverse package for importing tabular data.
fread()
- In the data.table package. Also imports .txt
files. Faster and more efficient than read.csv. Like
read.csv, the nrows and skip
inputs are particularly useful.
readxl -
A Tidyverse package for importing Excel files.
readLines() - A powerful but crude tool that dumps
the entire text content of files. Requires a lot of processing
afterwards to get any useful data. A last resort option.
janitor::clean_names()
- Handy for cleaning up column names, if you want to.
respR does not work with traditional date-time data
(e.g. “2023-03-12 13:46:55”). Instead the time data must be
in a numeric form in seconds (by far the most commonly used), minutes,
hours or days. The good news is that most export files from oxygen
probes systems contain a timestamp or numeric time column
already. If they do not, it is relatively easy to convert date-times to
a numeric format. See vignette("format_time").
import_file
Note as of respR v2.3.0 this function has
been DEPRECATED. It is still fully functional, but
import_file will not be updated and will be removed in a
future release. Users should move their importing data workflows to use
more generic functions such as those detailed above. This gives you much
more control and the ability to troubleshoot issues if a problem occurs,
rather than waiting for this function to be updated.
For files currently supported, import_file() uses
pattern recognition to identify the originating system of the file,
imports it, formats columns and column names, and outputs a data frame
that can be passed to respR functions.
To repeat: use of this function to prepare the data for use in
respR is OPTIONAL and in fact
NOT RECOMMENDED. It is a convenience function
and mainly intended for those completely new to R. If you
are even slightly experienced in R you should know how to
import data and structure it to a simple dataframe of time against
oxygen. It is ALWAYS better to import files yourself using
generic functions. See above.
import_file supports several systems at present,
including Firesting, Pyro, PreSens, MiniDOT, Loligo Witrox, Vernier and
more. See import_file() for full list.
If the imported data does not contain a numeric time or timestamp
column, but only dates and times see
vignette("format_time"). After your data is in a paired,
numeric time~oxygen structure, it can be passed to
inspect() to check the importing or time formatting has
worked, look for common errors, and visualise the dataset. See
vignette("inspecting").