| moss_select {MOSS} | R Documentation |
Returns features and subject selected by latent dimension.
Description
This function is meant to used after moss. Its main purpose is to extract the features and subjects by latent dimension. The selection depends on loadings at each dimension being different from zero.
Usage
moss_select(data.blocks, SVD, resp.block = NULL, K = NULL, plot = FALSE)
Arguments
data.blocks |
a list of omic blocks as provided to moss. |
SVD |
a list with SVD results. The function is meant to work with the results from sparse SVD. However, 'dense' solutions are also accepted. |
resp.block |
Which omic block was used as response in moss? Integer. Defaults to NULL. |
K |
How many dimensions should be displayed? Vector. Defaults to the 1 : ncol(SVD$v). |
plot |
Should the results be plotted? Logical. Defaults to FALSE |
Value
Returns a list containing the position, name, and loadings of selected features and subjects by latent dimension. if 'plot=TRUE', a scatterplot is displayed, where the x-axis represents the latent dimensions, the y-axis the total number of features selected in log scale, and each point is a pie chart showing the relative contribution of each omic to the number of features selected. The radio of the pie chart represents the coefficient of variation among squared loadings (mean squared loadings divided by their standard deviation).