mergeTrialInfo {servosphereR}

Merge trial id information

Description

This function merges trial id information with the servosphere data.

Usage

mergeTrialInfo(list, trial.data, col.names, stimulus.keep,
  stimulus.split = FALSE)

Arguments

list

The list of servosphere output data.

trial.data

The data frame containing your trial id information. This must contain an identifier column titled 'id' and if the data are to be split by stimuli, an additional identifier column 'id_stim'. See the description for more details.

col.names

A string vector containing the names of the columns you want to transfer to your servosphere output data. The trial.data data frame may have columns not needed for the analysis, so the function asks the user to be explicit about which columns to retain.

stimulus.keep

An integer vector containing the stimuli numbers to retained in the data and split the data frames by. Omitted stimuli values will be discarded.

stimulus.split

A logical value indicating whether the data frames should be split by stimulus. Defaults to 'FALSE'. If 'TRUE', be sure to include a 'id_stim' column to give each trial/stimulus combination a unique ID.

Details

Users of the servosphere will need a separate data frame with trial id information in a column titled id. This should contain a unique identifier and any other relevant experimental information, such as treatments applied, date, time of day, etc. Make sure the rows in your trial id data frame are ordered in the same order as the list of data frames of your servosphere output.

Researchers may wish to compare data before and after some stimulus has been applied and this function allows the user to split their data into separate data frames based on different values of the stimulus variable to facilitate these comparison. If the data frames should be split by stimulus, the trial record data frame should contain a column id_stim that lists the id number of the trial, followed by an underscore, followed by each value of the stimulus variable retained. If the data should be split by stimulus, the rows of the trial id data frame should be ordered first by id in the same order as their data is stored within the list. Once ordered by id the trial data data frame should be further ordered within an id by stimulus (i.e. id_stim 1_1 should come before id_stim 1_2).

Data recorded during a particular stimuli may also be discarded if it is not required for analysis. For example, recordings may begin with a five minute adjustment period and the data associated with that period may not be used for analysis. The stimulus recorded by the software during that five minute adjustment period can be discarded by omitting that stimulus number from the stimulus.keep argument.

This function will also append an item to your list of data frames that contains the relevant column names to be retained in future manipulations of the data.

Value

Returns the list of data frames provided which have been merged with additional relevant trial information.

Examples

 servosphere <- list(data.frame(id = rep(1, 200),
                                stimulus = rep(c(0, 1), each = 100),
                                dT = sample(8:12, 200, replace = TRUE),
                                dx = runif(200, 0, 5),
                                dy = runif(200, 0, 5),
                                treatment = rep("a", 200),
                                date = rep("2032018", 200)),
                     data.frame(id = rep(2, 200),
                                stimulus = rep(c(0, 1), each = 100),
                                dT = sample(8:12, 200, replace = TRUE),
                                dx = runif(200, 0, 5),
                                dy = runif(200, 0, 5),
                                treatment = rep("b", 200),
                                date = rep("2032018", 200)))
 trial_record <- data.frame(id = c(1, 2),
                            treatment = c("a", "b"),
                            date = c("2032018", "2042018"),
                            time = c("13:30", "07:30"))
 trial_id_split <- data.frame(id = c(1, 2, 1, 2),
                              stimulus = c(1, 1, 2, 2),
                              treatment = c("a", "b", "a", "b"),
                              date = rep(c("2032018", "2042018"), times = 2),
                              time = rep(c("13:30", "07:30"), times = 2),
                              id_stim = c("1_1", "2_1", "1_2", "2_2"))
# Merge the columns id, treatment, and date from the trial_record data frame
# with all the data frames in our list servosphere.

 merged_servosphere <- mergeTrialInfo(servosphere,
   trial_record,
   col.names = c("id", "treatment"),
   stimulus.keep = c(0, 1))

# Repeat of the merger above without retaining the id column while
# also splitting the data provided into separate data frames based on the
# different stimuli recorded, keeping only data associated with stimuli 1 and
# 2. Splitting based on stimulus requires a column in the trial.data data
# frame titled id_stim.

 merged_servosphere <- mergeTrialInfo(servosphere,
     trial_id_split,
     col.names = c("id", "treatment"),
     stimulus.split = TRUE,
     stimulus.keep = c(0, 1))