Use of the pre-computed LD matrices

[SalemWerdyani]

Hi Privé,

I am new to the LDperd2. It is a bit comfusing, following to match the target genetic data with the SumStat data, do I need to compute my LD matrices using the following code:

tmp <- tempfile(tmpdir = "file.size(corr$sbk) / 1024^3 ")

for (chr in 1:22) {
  
  # print(chr)
  
  ## indices in 'df_beta'
  ind.chr <- which(df_beta$chr == chr)
  ## indices in 'G'
  ind.chr2 <- df_beta$`_NUM_ID_`[ind.chr]
  
  corr0 <- snp_cor(G, ind.col = ind.chr2, size = 3 / 1000,
                   infos.pos = POS2[ind.chr2], ncores = NCORES)
  
  if (chr == 1) {
    ld <- Matrix::colSums(corr0^2)
    corr <- as_SFBM(corr0, tmp, compact = TRUE)
  } else {
    ld <- c(ld, Matrix::colSums(corr0^2))
    corr$add_columns(corr0, nrow(corr))
  }
}

I tried to run the above code for chr 22, but the system crashed.

If I want to use the pre-computed LD matrices, at what step can I use and what is the code?

Thank you,

[privefl]

In the LDpred2 tutorial, there is an example R script using the pre-computed LD matrices.

[SalemWerdyani]

Thank you, but it would be great if you can point to that example in your tutorial please.

[privefl]

https://privefl.github.io/bigsnpr/articles/LDpred2.html#which-set-of-variants-to-use