In ropensci/workloopR: Analysis of Work Loops and Other Data from Muscle Physiology Experiments

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

workloopR's data import functions, such as read_ddf(), are generally geared towards importing data from .ddf files (e.g. those generated by Aurora Scientific's Dynamic Muscle Control and Analysis Software).

Should your data be stored in another file format, you can use the as_muscle_stim() function to generate your own muscle_stim objects. These muscle_stim objects are used by nearly all other workloopR functions and are formatted in a very specific way. This helps ensure that other functions can interpret data & metadata correctly and also perform internal checks.

Load packages

Before running through anything, we'll ensure we have the packages we need.

library(workloopR)
library(magrittr)
library(ggplot2)

Data

Because it is somewhat difficult to simulate muscle physiology data, we'll use one of our workloop files, deconstruct it, and then re-assemble the data via as_muscle_stim().

## Load in the work loop example data from workloopR
workloop_dat <-
  system.file(
    "extdata",
    "workloop.ddf",
    package = 'workloopR') %>%
  read_ddf(phase_from_peak = TRUE) %>%
  fix_GR(GR = 2)

## First we'll extract Time
Time <- workloop_dat$Time
## Now Position
Position <- workloop_dat$Position
## Force
Force <- workloop_dat$Force
## Stimulation
Stim <- workloop_dat$Stim

## Put it all together as a data.frame
my_data <- data.frame(Time = Time,
                      Position = Position,
                      Force = Force,
                      Stim = Stim)

head(my_data)

Assemble via as_muscle_stim()

It is absolutely crucial that the columns be named "Time", "Position", "Force", and "Stim" (all case-sensitive). Otherwise, as_muscle_stim() will not interpret data correctly.

At minimum, this data.frame, the type of experiment, and the frequency at which data were recorded (sample_frequency, as a numeric) are necessary for as_muscle_stim().

## Put it together
my_muscle_stim <- as_muscle_stim(x = my_data,
                                 type = "workloop",
                                 sample_frequency = 10000)

## Data are stored in columns and basically behave as data.frames
head(my_muscle_stim)

ggplot(my_muscle_stim, aes(x = Time, y = Position)) +
  geom_line() + 
  labs(y = "Position (mm)", x = "Time (secs)") +
  ggtitle("Time course of length change") +
  theme_bw()

Attributes

By default, a couple attributes are auto-filled based on the available information, but it's pretty bare-bones

str(attributes(my_muscle_stim))

We highly encourage you to add in as many of these details as possible by passing them in via the ... argument. For example:

## This time, add the file's name via "file_id"
my_muscle_stim <- as_muscle_stim(x = my_data,
                                 type = "workloop",
                                 sample_frequency = 10000,
                                 file_id = "workloop123")

## For simplicity, we'll just target the file_id attribute directly instead of 
## printing all attributes again
attr(my_muscle_stim, "file_id")

Possible attributes

Here is a list of all possible attributes that can be filled.

names(attributes(workloop_dat))

To see how each should be formatted, (e.g. which ones take numeric values vs. character vectors...etc)

str(attributes(workloop_dat))

Thanks for reading!

Please feel free to contact either Vikram or Shree with suggestions or code development requests. We are especially interested in expanding our data import functions to accommodate file types other than .ddf in future versions of workloopR.

ropensci/workloopR documentation built on July 21, 2024, 4:31 p.m.