unifiedWMWqPCR (version 1.8.0)

uWMW,matrix-method: The unified Wilcoxon-Mann-Whitney test for qPCR data

Description

This function carries out the unified Wilcoxon-Mann-Whitney test for qPCR data. See De Neve et al. (2013) for more details.

Usage

"uWMW"(x, groups, housekeeping.names = NULL, transpose = FALSE, feat.names = NULL)
"uWMW"(x, feat.names = NULL, ...)
"uWMW"(x, groups, feat.names, subjects, value, ...)

Arguments

x

An object containing the qPCR measurements. See details.

groups

A vector indicating the groups that need comparing, or a single character telling which variable in the data frame contains the groups. Make sure this vector is as long as the number of replicates in the data set.

feat.names

An (optional) character vector with the names of the features (typically genes or microRNAs) or a single character giving the name of the feature variable. If not specified, the feature names are derived from the row names of the matrix, or from the feature names of the qPCRset object.

subjects

An (optional) character string indicating which variable of the data frame contains the subject id's. Ignored if x is not a data frame.

value

An (optional) character string indicating which variable of the data frame contains the values. Ignored if x is not a data frame.

housekeeping.names

an (optional) vector with the names of one or more housekeeping features. Make sure those names are spelled exactly as in the object.

transpose

In case a matrix is used, should the matrix be transposed? A matrix needs to be transposed when the columns do not represent the replicates. The function expects the columns to be replicates and the rows to be the different features.

...

For passing arguments between methods and to internal functions.

Value

An object of the class uwmwRes, containing the results of the unified Wilcoxon-Mann-Whitney test. See the help page of the class uwmwRes for more information.

Details

This function carries out the unified Wilcoxon-Mann-Whitney test using either Overall normalization (O) or Housekeeping normalization (H) as reference (see De Neve et al., 2013). If the argument housekeeping.names is specified, housekeeping normalization is considered. Otherwise overall normalization is considered.

The function uWMW can deal with data frames, matrices and qPCRset objects from the package HTqPCR. When using a data frame, you need to specify the arguments groups, feat.names, subjects and value; each one should contain the name of the related variable in the data frame.

When using a matrix, each column is assumed to be a subject and each row a feature. The argument groups should contain as much values as there are columns in the matrix. Note that the method for qPCRset objects does nothing else but extract the data matrix and do the analysis. You still need to specify where the groups are to be found.

References

De Neve, J. Thas, O. Ottoy, J.P. and Clement L. (2013) An extension of the Wilcoxon-Mann-Whitney test for analyzing RT-qPCR data. Statistical Applications in genetics and Molecular Biology. 12, 333-346

See Also

uwmw_Accessors and uwmw_Extract for accessing the results, and volcanoplot and forestplot for plotting them.

Examples

data(NBmat)
NBtest <- uWMW(NBmat, groups=NBgroups)

data(NBdata)
NBtest <- uWMW(x = NBdata, groups = "group", sub="subject",feat="miRNA",val="Cq")
head(NBtest)
as.matrix(NBtest)