View source: R/clean_obs_data.R
clean_obs_data | R Documentation |
The function clean_obs_data
is meant to perform the following
automated data cleaning steps:
Make all the main columns that you probably want in concentration-time data – Subject, Day, Time, DV, Conttime, etc. – be the 1st columns listed and have standardized names.
Remove "mg" or "mg/kg" or "mg/m2" from any column with the dose info.
If there's a column for the day and a column for the time, calculate the continuous time.
Set any times < 0 to 0.
Split the data into separate csv files or list items by user-specified columns.
A note:
When you specify which columns are which in the arguments, the names must
EXACTLY match what R reads in for an Excel file. This can be tricky if there
are any spaces or especially any carriage returns in the column headings in
an Excel file. Put tick marks around any column names with spaces or special
characters, e.g., 'Subject ID'. clean_obs_data
will not do
everything you'll probably need since messy data are messy in their
own unique ways in every case, but it will make a start. Once your observed
concentration-time data are in good shape, we recommend checking out
format_obs_for_XML
for getting the data into the shape
necessary for pasting into a Simcyp Simulator PE template Excel file,
including adding dosing rows automatically based your specifications for the
dosing interval.
clean_obs_data(
untidy_data,
untidy_data_sheet = NA,
cohort_column,
subject_column,
time_column,
DV_column,
analyte_column,
dose_column,
day_column,
age_column,
weight_column,
height_column,
sex_column,
SDSE_column,
injection_site_column,
dose_unit = "mg",
weighting = 1,
split_columns = NA,
save_csv = NA
)
untidy_data |
a data.frame to be tidied or a csv file with the untidy data or an Excel file with the untidy data. This should be in quotes. |
untidy_data_sheet |
if you supplied an Excel file with the untidy data, specify what sheet to read here. This should be in quotes. |
cohort_column |
the column in the observed concentration-time data.frame that contains the cohorts, unquoted |
subject_column |
the column in the observed concentration-time data.frame that contains the subject IDs, unquoted. |
time_column |
the column in the untidy data.frame that contains times, unquoted. |
DV_column |
the column in the untidy data.frame that contains concentration data, unquoted. |
analyte_column |
the column in the untidy data.frame that contains the analytes, unquoted. |
dose_column |
the column in the untidy data.frame that contains the doses, unquoted. |
day_column |
the column in the untidy data.frame that contains the days, unquoted. |
age_column |
the column in the untidy data.frame that contains the ages, unquoted. |
weight_column |
the column in the untidy data.frame that contains the weights, unquoted. |
height_column |
the column in the untidy data.frame that contains heights, unquoted. |
sex_column |
the column in the untidy data.frame that contains sexes, unquoted. |
SDSE_column |
the column in the untidy data.frame that contains the standard deviation or standard error, unquoted. |
injection_site_column |
the column in the untidy data.frame that contains the injection site, unquoted. |
dose_unit |
the unit of dosing. Options are "mg" (default), "mg/m2", or "mg/kg". |
weighting |
weighting to use for parameter estimation. Defaults to 1. |
split_columns |
Which columns should the data be split up by? This should be quoted. Apologies for the difference here since the other column assignments are unquoted, but, since this can have multiple columns, the coding is easiest to do if these are in quotes. We're trying to figure out how to get around that. |
save_csv |
optionally specify a file name for saving the tidied data as a csv. If you have more than one dose level in the data, the files will be split up by dose level, and we'll include a suffix with the dose amount at the end of the file name. |
a list of tidied data, split by dose level
# none yet
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.