- mo_data_target
-Symbol, the name of the target containing the MultiDataSet
object.
+Symbol, the name of the target containing the
+MultiDataSet
object.
- transformations
-Named character vector, name of each element is the name of a dataset
-to transform, corresponding element gives the type of transformation to apply to the dataset
-(e.g. c(rnaseq = 'vst-deseq2', phenotypes = 'best-normalize-auto')
). See Details
-for a list of available transformations. If 'best-normalize-auto'
is selected, need to
-provide the methods
argument as well.
+Named character vector, name of each element is the
+name of a dataset to transform, corresponding element gives the type of
+transformation to apply to the dataset (e.g. c(rnaseq = 'vst-deseq2', phenotypes = 'best-normalize-auto')
). See Details for a list of available
+transformations. If 'best-normalize-auto'
is selected, need to provide
+the methods
argument as well.
- return_matrix_only
-Logical, should only the transformed matrix be returned for each
-transformation? If TRUE
, only transformed matrices will be stored. If FALSE
,
-instead for each transformation, a list with the transformed data and potentially other
-information relevant to the transformation will be saved. Default value is FALSE
.
+Logical, should only the transformed matrix be
+returned for each transformation? If TRUE
, only transformed matrices will
+be stored. If FALSE
, instead for each transformation, a list with the
+transformed data and potentially other information relevant to the
+transformation will be saved. Default value is FALSE
.
- target_name_prefix
-Character, a prefix to add to the name of the targets created by this target factory.
-Default value is ""
.
+Character, a prefix to add to the name of the
+targets created by this target factory. Default value is ""
.
- transformed_data_name
-Character, the name of the target containing the MultiDataSet
with
-transformed data to be created. If NULL
, will be selected automatically. Default value is NULL
.
+Character, the name of the target containing the
+MultiDataSet
with transformed data to be created. If NULL
, will be
+selected automatically. Default value is NULL
.
+
+
+- log_bases
+Numeric or named numeric list, gives for each dataset for
+which the 'logx'
transformation is selected the log base to use. If one
+value, will be used for all concerned datasets. Otherwise, can specify a
+different log-base for each concerned dataset by passing a named list.
+
+
+- pre_log_functions
+Function or named list of functions, gives for each
+dataset for which the `'logx`` transformation is selected the function that
+will be applied to the matrix before the log transformation (e.g. to apply
+an offset to the values to avoid issues with zeros). Default value is the
+zero_to_half_min()
function. If one value, will be used for all concerned
+datasets. Otherwise, can specify a different log-base for each concerned
+dataset by passing a named list.
- methods
-Named character vector, gives for each dataset for which the 'best-normalize-manual'
-transformation is selected which normalisation method should be applied. See possible values in
-Details.
+Character or named character list, gives for each dataset for
+which the 'best-normalize-manual'
transformation is selected the
+normalisation method that should be applied. See possible values in
+Details. If one value, will be used for all concerned datasets. Otherwise,
+can specify a different method for each concerned dataset by passing a
+named list.
- ...
-Further arguments passed to the transform_dataset
function or the
-method
function from the bestNormalize
package. Only relevant for 'best-normalize-XX'
transformations.
+Further arguments passed to the transform_dataset
+function or the method
function from the bestNormalize
package. Only
+relevant for 'best-normalize-XX'
transformations.