scale of MCMC sampling of beta

[yidingdd]

Hi Florian,

Could you help to confirm the scale of betas from the MCMC sampling in your newest bigsnpr release?

It looks like the beta sample returned by snp_ldpred2_grid is on allelic scale, but the beta sample returned by snp_ldpred2_auto is on standardized scale, is it correct?

bigsnpr/R/LDpred2.R

Line 139 in ee9f2ed

sweep(beta_gibbs, 1, scale, '*')

bigsnpr/R/LDpred2.R

Line 182 in ee9f2ed

#' - `$sample_beta`: sparse matrix of sampling betas (see parameter `report_step`),

Best,
Yi

[privefl]

Hi Yi,

For LDpred2-auto,

• $beta_est is also on allelic scale
• but $sample_beta is indeed on the standardized scale. You can probably use Matrix::Diagonal(x = scale) %*% auto$sample_beta to transform it to the allelic scale, where scale <- with(df_beta, sqrt(n_eff * beta_se^2 + beta^2)).

[yidingdd]

Hi Florian,

Thank you for confirming LDpred2-auto, for ldpred2-grid, do we need to use the same rescaling?

Best,
Yi

[privefl]

No, the output is already on the allelic scale for LDpred2-grid.