Extract weights from sdpr function

Usage

sdprWeights(
  stat,
  ldReference,
  fitRetention = c("none", "slim", "full"),
  methodArgs = SdprOptions()
)

Arguments

stat: A list of summary statistics: n (per-variant sample sizes) and the marginal effects as cor (standardized), z, b with seb, or b alone, taken as already standardized. The engine receives the standardized effects z / sqrt(n), with n the median of n.
ldReference: The LD reference the statistics' variants are matched to by stat$variantNames: a genotype panel (see readGenotypes), a GenotypeMixture or an LdMatrix. A genotype reference's LD is used through its standardized genotypes and not formed as a matrix, except under optLlk = 2 when the per-variant sample sizes and arrays are too varied for its low-rank form to pay.
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 sampler's estimates, betaEst and the heritability h2. This engine keeps no trimmable intermediates, so the two retaining levels behave alike here.
methodArgs: Options forwarded to the sampler, sdpr, built with SdprOptions. 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 the posterior SNP coefficients. Unless fitRetention = "none", the sampler's estimates, the heritability among them, are attached as attribute "fit".

Examples

data(eqtlRegionExample)
X <- eqtlRegionExample$X[, 1:30]
y <- eqtlRegionExample$yRes
ss <- lapply(
  seq_len(ncol(X)), function(j) coef(summary(lm(y ~ X[, j])))[2, 1:2])
stat <- list(
  b = vapply(ss, `[`, numeric(1), 1L),
  seb = vapply(ss, `[`, numeric(1), 2L),
  n = rep(nrow(X), ncol(X)),
  variantNames = colnames(X)
)
w <- sdprWeights(stat, LdMatrix(cor(X)), fitRetention = "slim")
#> 100 iter. h2: 0.0208449 max beta: 0.0411226
#> 200 iter. h2: 0.00251109 max beta: 0.0286469
#> 300 iter. h2: 4.4499e-05 max beta: 0.00135623
#> 400 iter. h2: 0.00107698 max beta: -0.0164433
#> 500 iter. h2: 0.000290251 max beta: 0.00942079
#> 600 iter. h2: 0 max beta: -0
#> 700 iter. h2: 0 max beta: 0
#> 800 iter. h2: 0 max beta: -0
#> 900 iter. h2: 0 max beta: -0
#> 1000 iter. h2: 0 max beta: 0
#> h2: 0.00277866 max: 0.00262481
attr(w, "fit")$h2
#> [1] 0.002778656