Get Correlations Between CSs, using Variable with Maximum PIP From Each CS

Source: R/susie_utils.R

This function evaluates the correlation between single effect CSs. It is not part of the SuSiE inference. Rather, it is designed as a diagnostic tool to assess how correlated the reported CS are.

get_cs_correlation(model, X = NULL, Xcorr = NULL, max = FALSE)

Arguments

model

A SuSiE fit, typically an output from susie or one of its variants.

X

n by p matrix of values of the p variables (covariates) in n samples. When provided, correlation between variables will be computed and used to remove CSs whose minimum correlation among variables is smaller than min_abs_corr.

Xcorr

p by p matrix of correlations between variables (covariates). When provided, it will be used to remove CSs whose minimum correlation among variables is smaller than min_abs_corr.

max

When max = FAFLSE, return a matrix of CS correlations. When max = TRUE, return only the maximum absolute correlation among all pairs of correlations.

Value

A matrix of correlations between CSs, or the maximum absolute correlation when max = TRUE.