OPM: Real classes of the 'opm' package
In opm: Analysing Phenotype Microarray and Growth Curve Data

Classes whose members can be generated and manipulated by an opm user: OPM, OPMA, OPMD, OPMS and MOPMX.

OPM is an acronym for ‘OmniLog(R) Phenotype Microarray’. This is the class for holding single-plate OmniLog(R) phenotype microarray data without aggregated values, but with information read from the original input CSV files as well as an additional arbitrary amount of arbitrarily organised metadata. Objects of this class are usually created by inputting files with read_single_opm or read_opm, not with a call to new or as.

OPM inherits from WMD and, hence, has all its methods.

Regarding the coercion of this class to other classes (see the as method from the methods package), consider the following:

Conversion of its child classes to this class is straightforward, see the examples below.
The coercion of this class (and its child classes) to a list (and vice versa) is only for expert users and relies on a mapping between slot names and keys in the list, i.e. the list must be appropriately named. For instance, this is the mechanism when reading from and writing to YAML, see to_yaml.
Coercions to other data frames and matrices first coerce the measurements and then add the other slots as attributes.
Methods such as flatten and extract might be way more appropriate for converting OPM objects.

OPMA is an acronym for ‘OPM, aggregated’. This is the class for holding single-plate OmniLog(R) phenotype microarray data together with aggregated values. Objects of this class are usually created by calling do_aggr on an OPM object, or by inputting files with read_single_opm or read_opm if these files already contain aggregated data.

OPMA inherits from OPM and, hence, has all its methods.

OPMD is an acronym for ‘OPM, discretised’. This is the class for holding single-plate OmniLog(R) phenotype microarray data together with aggregated and discretised values. Objects of this class are usually created by calling do_disc on an OPMA object, or by inputting files with read_single_opm or read_opm if these files already contain discretised data.

OPMD inherits from OPMA and, hence, has all its methods.

The discretised data are considered as ‘consistent’ with the curve parameter from which they have been estimated if no FALSE value corresponds to curve parameter larger than the curve parameter of any TRUE value; NA values are not considered when checking consistency. The strict.OPMD entry of opm_opt determines whether an error or only a warning is issued in the case of inconsistency.

OPMS is the class for holding multiple-plate OmniLog(R) phenotype microarray data with or without aggregated or discretised values. Regarding the name: OPMS is just the plural of OPM. Objects of this class are usually created by calling opms or other combination functions on OPM or derived objects, or by inputting files with read_opm if these files altogether contain more than a single plate. OPMS objects are not normally created with a call to new or as. The data may have been obtained from distinct organisms and/or replicates, but must correspond to the same plate type and must contain the same wells.

OPMS inherits from WMDS and, hence, has all its methods. As a rule, OPMS has the same methods as the OPM class, but adapted to a collection of more than one OPM object. Also, OPMS can hold OPMD and OPMA as well as OPM objects, even though this is not indicated for all its methods in this manual.

MOPMX is an object for holding OPMX objects with potentially multiple plate types. Regarding the name: the M stands for ‘multiple’. MOPMX objects are generated by read_opm(convert = "grp") and opms(group = TRUE). MOPMX objects can also be created with new or as and then further manipulated; see the examples below. MOPMX objects in many aspects behave like lists.

OPM_MCP_OUT is a data-frame based class useful as intermediate result of opm_mcp. See there and its annotated method for usages.

methods::Methods methods::new

Other classes: FOE, OPMA_DB, OPMD_DB, OPMX, OPM_DB, WMD, WMDS, WMDX, XOPMX, YAML_VIA_LIST

## overview on the classes
showClass("OPM")
showClass("OPMA")
showClass("OPMD")
showClass("OPMS")
showClass("MOPMX")

## OPMX conversions with as()
showMethods("coerce", classes = c("OPM", "OPMA", "OPMD", "OPMS"))
data(vaas_1)
data(vaas_4)
(x <- as(vaas_1, "OPMA")) # drops the discretised data
stopifnot(has_disc(vaas_1), !has_disc(x))
(x <- as(vaas_1, "OPM")) # drops the aggregated data
stopifnot(has_aggr(vaas_1), !has_aggr(x))

## MOPMX creation and conversion
(x <- new("MOPMX")) # don't do this with the other classes
(x <- as(vaas_1, "MOPMX"))
(x <- as(vaas_4, "MOPMX"))
# conversion backwards is only possible as long as the MOPMX object contains
# only a single OPMX object
showMethods("coerce", classes = "MOPMX")