amData {AMModels}R Documentation

Create An amData Object That Pairs Datasets With Associated Metadata


Creates an object of class amData, which is typically a data frame of covariate data or model fitting data, with mandatory metadata. It is worth noting that some models include the data as part of the fitted model object, e.g. lm and glm, which access embedded data with model.frame. For large datasets that are embedded in the model, it may be worth documenting metadata a placeholder object such as the character string "embedded data" rather than a redundant data object.

An S4 class to store data with descriptive metadata.


amData(data, ...)



A dataset, typically data frame but may be in any structure.


Named metadata elements, either supplied individually and coerced to list or supplied as a named list.


An object of class amData suitable for inclusion in an amModelLib object.



A single object containing data relevant to a model. There are no restrictions on how the data are used in the model; for example, they may be covariate data (for use on the right side of the equation) or observed data (for use on the left side of the equation).


A named list of length 1 character vectors that form name:value pairs, e.g the source, the collection method, etc. When embedded in an amModelLib, data metadata may be retrieved or set with dataMeta.

See Also

Other amModelLib: AMModels, amModelLib, amModel, getters, grepAMModelLib, insertAMModelLib, lsModels, rmModel


# create dataset from lm helpfile
## Annette Dobson (1990) "An Introduction to Generalized Linear Models".
## Page 9: Plant Weight Data.
ctl <- c(4.17,5.58,5.18,6.11,4.50,4.61,5.17,4.53,5.33,5.14)
trt <- c(4.81,4.17,4.41,3.59,5.87,3.83,6.03,4.89,4.32,4.69)
group <- gl(2, 10, 20, labels = c("Ctl","Trt"))
weight <- c(ctl, trt)

# create a dataset that is of class data.frame <- data.frame(weight, group)

# create an amData data object 
dat1 <- amData(data =, comment='Dataset from lm helpfile.', taxa = 'plants')

# the class of dat1 is amData

# the summary function will invoke the summary method for the dataset's original class

# use the amModelLib function to create a new amModelLib called mymodels that 
# includes dat1; data must be supplied in a named list
mymodels <- amModelLib(
    data=list(dat1 = dat1), 
    description = "An example amModelLib called mymodels."

# use the lsData function to list the amData objects in an amModelLib

# the dataMeta function can be used to retrieve an amData object's metadata
dataMeta(amml = mymodels, 'dat1')

#  the dataMeta function can alse be used to set metadata
dataMeta(mymodels, 'dat1') <- list(
    url = ""
dataMeta(amml = mymodels, 'dat1')

# use the getAMData function to extract the dataset back to its orginal form
getAMData(amml = mymodels, 'dat1', as.list = FALSE)

# the retrieved datset is in its original class
class(getAMData(amml = mymodels, 'dat1', as.list = FALSE))

# use the amModelLib function to create an empty amModelLib
mymodels2 <- amModelLib(description = "An example amModelLib called mymodels2.")

# use the insertAMModelLib function to insert the amData object to an 
# existing amModelLib
mymodels2 <- insertAMModelLib(data = list(dat1 = dat1))

# use rmData to remove an amData object from an amModelLib
rmData('dat1', amml = mymodels2)

[Package AMModels version 0.1.4 Index]