weitrix_calibrate: Adjust weights row-wise based on given row dispersions

In pfh/weitrix: Tools for matrices with precision weights, test and explore weighted or sparse data

Adjust weights row-wise based on given row dispersions

Description

Based on estimated row dispersions, adjust weights in each row.

Usage

weitrix_calibrate(weitrix, dispersions)

Arguments

weitrix	A weitrix object, or an object that can be converted to a weitrix with as_weitrix.
dispersions	A dispersion for each row.

Details

For large numbers of samples this can be based directly on weitrix_dispersions. For small numbers of samples, when using limma, it should be based on a trend-line fitted to known co-variates of the dispersions. This can be done using weitrix_calibrate_trend.

Value

A SummarizedExperiment object with metadata fields marking it as a weitrix.

Examples

# Adjust weights so dispersion for each row is exactly 1. This is dubious 
# for a small dataset, but would be fine for a dataset with many columns.
comp <- weitrix_components(simwei, p=1, verbose=FALSE)
disp <- weitrix_dispersions(simwei, comp)
cal <- weitrix_calibrate(simwei, disp)
weitrix_dispersions(cal, comp)

pfh/weitrix documentation built on Dec. 10, 2024, 5:56 p.m.