Compute Weights Using mr.ash Shrinkage

This function fits the `mr.ash` model (adaptive shrinkage regression) to estimate weights for a given set of predictors and response. The prior is susieR::mr.ash's own default mixture grid unless sa2 is supplied through methodArgs, and custom initial beta values are accepted.

Usage

mrashWeights(
  X,
  y,
  fitRetention = c("none", "slim", "full"),
  methodArgs = MrashOptions()
)

Arguments

X

Numeric genotype / design matrix (samples x variants).

y

Numeric response (phenotype) vector of length nrow(X).

fitRetention

How much of the fit is kept as the "fit" attribute of the returned weights: "none" (default) attaches nothing, "slim" and "full" attach the fitted model. This engine keeps no trimmable intermediates, so the two retaining levels behave alike here.

methodArgs

Options forwarded to susieR::mr.ash, built with MrashOptions. A bare list is refused: it cannot be checked against the engine, so a misspelled option would be silently ignored.

Value

A numeric vector of weights, one per variant (column of X); zero-variance columns receive weight 0. Unless fitRetention = "none" the fitted mr.ash object is attached as attribute "fit".

Examples

set.seed(1)
X <- matrix(rnorm(200 * 10), nrow = 200)
y <- as.numeric(X %*% rnorm(10) + rnorm(200))
weights <- mrashWeights(X, y)
#> Mr.ASH terminated at iteration 18: max|beta|=2.1773e+00, sigma2=1.2272e+00, pi0=0.0111
#> Warning: The mixture proportion associated with the largest prior variance is greater than 4.00e-02; this indicates that the model fit could be improved by using a larger setting of the prior variance. Consider increasing the range of the variances "sa2".