as.data.frame: Create data frame
In opm: Analysing Phenotype Microarray and Growth Curve Data

Description Usage Arguments Details Value See Also Examples

These as.data.frame methods create a data frame from aggregated and discretised values in a manner distinct from extract. flatten converts into a ‘flat’ data frame, including all measurements in a single column (suitable, e.g., for lattice).

  ## S4 method for signature 'MOPMX'
as.data.frame(x, row.names = NULL,
    optional = FALSE, sep = "_", csv.data = TRUE, settings = TRUE,
    include = FALSE, ..., stringsAsFactors = default.stringsAsFactors()) 
  ## S4 method for signature 'OPM'
as.data.frame(x, row.names = NULL,
    optional = FALSE, sep = "_", csv.data = TRUE, settings = TRUE,
    include = FALSE, ..., stringsAsFactors = default.stringsAsFactors()) 
  ## S4 method for signature 'OPMA'
as.data.frame(x, row.names = NULL,
    optional = FALSE, sep = "_", csv.data = TRUE, settings = TRUE,
    include = FALSE, ..., stringsAsFactors = default.stringsAsFactors()) 
  ## S4 method for signature 'OPMD'
as.data.frame(x, row.names = NULL,
    optional = FALSE, sep = "_", csv.data = TRUE, settings = TRUE,
    include = FALSE, ..., stringsAsFactors = default.stringsAsFactors()) 
  ## S4 method for signature 'OPMS'
as.data.frame(x, row.names = NULL,
    optional = FALSE, sep = "_", csv.data = TRUE, settings = TRUE,
    include = FALSE, ..., stringsAsFactors = default.stringsAsFactors()) 
  ## S4 method for signature 'kegg_compound'
as.data.frame(x, row.names = NULL,
    optional = TRUE, ..., stringsAsFactors = FALSE) 
  ## S4 method for signature 'kegg_compounds'
as.data.frame(x, row.names = NULL,
    optional = TRUE, ..., stringsAsFactors = FALSE) 

  ## S4 method for signature 'MOPMX'
flatten(object, include = NULL, fixed = list(),
    factors = FALSE, ...) 
  ## S4 method for signature 'OPM'
flatten(object, include = NULL, fixed = list(),
    factors = TRUE, exact = TRUE, strict = TRUE, full = TRUE,
    numbers = FALSE, ...) 
  ## S4 method for signature 'OPMS'
flatten(object, include = NULL, fixed = list(), ...)

`x`	Object of class `OPM`, its child classes, or `OPMS` or `MOPMX`. If an `OPMS` object, for the `as.data.frame` method its elements must either all be `OPM` or all be `OPMA` or all be `OPMD` objects. If a `MOPMX` object, its elements must be conforming `OPMS` or either `OPM`, `OPMA` or `OPMS` objects. There are `as.data.frame` methods for some of the objects created by `substrate_info`, too.
`row.names`	Optional vector for use as row names of the resulting data frame. Here, it is not recommended to try to set row names explicitly.
`optional`	Logical scalar passed to the list and matrix methods of `as.data.frame`.
`sep`	Character scalar used as word separator in column names. Set this to `NULL` or an empty vector to turn off character replacement in column names.
`csv.data`	Logical scalar indicating whether the `csv_data` entries that identify the plate shall be included.
`settings`	Logical scalar indicating whether the `aggr_settings` and `disc_settings` entries, if available, shall be included.
`stringsAsFactors`	Logical scalar passed to the list and matrix methods of `as.data.frame`.
`object`	`OPM` or `OPMS` object (or list).
`include`	For `flatten`, either `NULL`, character vector, list or formula. If not empty, include this meta-information in the data frame, replicated in each row. Otherwise it converted to a list and passed to `metadata`. See there for details. For `as.data.frame`, if empty or `FALSE`, ignored. If `TRUE`, all metadata are included using `to_metadata`. If otherwise and non-empty, metadata selected using `extract_columns` are included.
`fixed`	`NULL` or list. If not `NULL`, include these items in the data frame, replicated in each row.
`factors`	Logical scalar. See the `stringsAsFactors` argument of `data.frame` and `as.data.frame` from the base package.
`exact`	Logical scalar. Passed to `metadata`.
`strict`	Logical scalar. Passed to `metadata`.
`full`	Logical scalar. Replace well coordinates by full names?
`numbers`	Logical scalar. Use numbers instead of well names? This is not recommended for must usages.
`...`	Optional other arguments passed to `wells`, or from the `OPMS` to the `OPM` method, or to the list and matrix methods of `as.data.frame`.

The as.data.frame methods for OPMX objects are mainly intended to produce objects that can easily be written to CSV files, for instance using write.table from the utils package. There are no opm methods other than batch_opm (which can write such files) that make use of the created kind of objects. In particular, they cannot be input again into opm.

The following entries are contained in the generated data frame:

Optionally the csv_data entries that identify the plate.
The names of the wells. Always included.
For OPMA objects (and OPMS objects that contain them as well as MOPMX objects that contain such OPMA or OPMS objects), always the aggregated data (curve parameters), one column for each point estimate, upper and lower confidence interval of each parameter.
For OPMA objects (and OPMS objects that contain them as well as MOPMX objects that contain such OPMA or OPMS objects), optionally the used aggregation settings, one column per entry, except for the ‘options’ entry (which is not a scalar). The column names are prefixed with "Aggr" followed by sep. If sep is empty, opm_opt("comb.key.join") is used.
For OPMD objects (and OPMS objects that contain them as well as MOPMX objects that contain such OPMD or OPMS objects), always one column with the discretised data.
For OPMD objects (and OPMS objects that contain them as well as MOPMX objects that contain such OPMD or OPMS objects), optionally the used discretisation settings, one column per entry, except for the ‘options’ entry (which is not a scalar). The column names are prefixed with "Disc" followed by sep. If sep is empty, opm_opt("comb.key.join") is used.

The limits of using CSV as output format already show up in this list, and in general we recommend to generate YAML or JSON output instead.

For the as.data.frame methods of the other classes, see substrate_info.

In the data frame returned by flatten, column names are unchecked (not converted to variable names). The three last columns are coding for time, well and value, with the exact spelling of the column names given by param_names.

The OPMS method yields an additional column for the plate, the exact spelling of its name also being available via param_names. This column contains the position of each plate within object.

The MOPMX method yields a another additional column for the plate type. There is currently no safeguard against having several OPMX objects of the same plate type within a MOPMX object.

The as.data.frame methods create a data frame with one row for each combination of well and plate.

The flatten methods create a data frame with one row for each combination of time point, well and plate.

utils::write.table stats::reshape pkgutils::flatten

Other conversion-functions: extract, extract_columns, merge, oapply, opmx, plates, rep, rev, sort, split, to_yaml, unique

## OPMD method of as.data.frame()
summary(x <- as.data.frame(vaas_1))
stopifnot(is.data.frame(x), nrow(x) == 96)

## OPMS method of as.data.frame()
summary(x <- as.data.frame(vaas_4[, , 1:10]))
stopifnot(is.data.frame(x), nrow(x) == 10 * 4)

## OPM method of flatten()
# distinct numbers of columns due to distinct selection settings
head(x <- flatten(vaas_1))
stopifnot(is.data.frame(x), identical(dim(x), c(36864L, 3L)))
head(x <- flatten(vaas_1, fixed = "TEST", include = "Strain"))
stopifnot(is.data.frame(x), identical(dim(x), c(36864L, 5L)))

## OPMS method of flatten()
# distinct numbers of columns due to distinct selection settings
head(x <- flatten(vaas_4[, , 1:10]))
stopifnot(is.data.frame(x), identical(dim(x), c(15360L, 4L)))
head(x <- flatten(vaas_4[, , 1:10], fixed = "TEST", include = ~ Strain))
stopifnot(is.data.frame(x), identical(dim(x), c(15360L, 6L)))