sLDSC Postprocessing Pipeline

Postprocess polyfun's per-trait sLDSC outputs (already loaded into an SldscData object) into a single results object with per-trait tau*, EnrichStat with back-solved jackknife SE, and a DerSimonian-Laird random-effects meta-analysis across traits. All file I/O is done up front by the reader functions (readSldscAnnot, readSldscFrq, readSldscTrait); this pipeline is pure computation over the in-memory SldscData.

Usage

sldscPostprocessingPipeline(
  sldscData,
  mafCutoff = 0.05,
  targetCategories = NULL,
  targetLabels = NULL
)

Arguments

sldscData: An SldscData object bundling the annotation table, the reference-panel allele frequencies, and the per-trait single/joint polyfun runs.
mafCutoff: Numeric MAF cutoff applied via the object's frq table. Default 0.05. Set to 0 to opt out (requires frq data when > 0).
targetCategories: Optional character vector of target annotation names to retain. Auto-detected from the joint run (or first single run) when NULL.
targetLabels: Optional display names, same length / order as targetCategories, applied to every output column / tau* block colname.

Value

A list with per_trait (per-trait standardised tables), meta tables (tauStar, enrichment, enrichstat), and a params record of the call options.

Examples

mkRun <- function(cats) {
  n <- length(cats)
  list(categories = cats, tau = setNames(rep(1e-7, n), cats),
    tauSe = setNames(rep(3e-8, n), cats),
    enrichment = setNames(rep(2, n), cats),
    enrichmentSe = setNames(rep(0.4, n), cats),
    enrichmentP = setNames(rep(0.01, n), cats),
    propH2 = setNames(rep(0.2, n), cats),
    propSnps = setNames(rep(0.1, n), cats), h2g = 0.3,
    annotationJackknifeCoefs = matrix(1e-7 * seq(0.9, 1.1, length.out = 10),
  10, n, dimnames = list(NULL, cats)),
    nBlocks = 10L)
}
annot <- data.frame(CHR = c(1, 1, 1, 2, 2, 2), SNP = paste0("rs", 1:6),
  annot_A = c(1, 0, 1, 0, 1, 0), annot_B = c(2.1, 1.8, 2.5, 1.9, 2.3, 2))
frq <- data.frame(CHR = c(1, 1, 1, 2, 2, 2), SNP = paste0("rs", 1:6),
  MAF = rep(0.2, 6))
mkTrait <- function() {
  list(single = list(mkRun(c("annot_A_0", "baselineLD_0")),
    mkRun(c("annot_B_0", "baselineLD_0"))),
    joint = mkRun(c("annot_A_0", "annot_B_0", "baselineLD_0")))
}
traits <- setNames(list(mkTrait(), mkTrait()), c("traitX", "traitY"))
sldscData <- SldscData(annotData = annot, frqData = frq, traits = traits)
sldscPostprocessingPipeline(sldscData = sldscData)
#> [sldsc] Computing M_ref...
#> [sldsc]   M_ref = 6 (MAF cutoff 0.05)
#> [sldsc] Computing per-annotation sd...
#> [sldsc]   sd computed for 2 annotation columns
#> [sldsc] Detecting binary vs continuous annotations...
#> [sldsc] Auto-detected 2 target categories
#> [sldsc] Detected 1 baseline annotations: baselineLD_0
#> [sldsc] Standardizing 2 traits...
#> [sldsc] Running random-effects meta across traits...
#> $per_trait
#> $per_trait$traitX
#> $per_trait$traitX$summary
#> # A tibble: 2 × 20
#>   target    isBinary tauSingle tauSeSingle tauStarSingle tauStarSeSingle
#>   <chr>     <lgl>        <dbl>       <dbl>         <dbl>           <dbl>
#> 1 annot_A_0 TRUE     0.0000001  0.00000003   0.00000115      0.000000221
#> 2 annot_B_0 FALSE    0.0000001  0.00000003   0.000000577     0.000000111
#> # ℹ 14 more variables: enrichmentSingle <dbl>, enrichmentSeSingle <dbl>,
#> #   enrichmentPSingle <dbl>, enrichstatSingle <dbl>, enrichstatSeSingle <dbl>,
#> #   tauJoint <dbl>, tauSeJoint <dbl>, tauStarJoint <dbl>, tauStarSeJoint <dbl>,
#> #   enrichmentJoint <dbl>, enrichmentSeJoint <dbl>, enrichmentPJoint <dbl>,
#> #   enrichstatJoint <dbl>, enrichstatSeJoint <dbl>
#> 
#> $per_trait$traitX$tau_star_blocks_single
#>          annot_A_0    annot_B_0
#>  [1,] 1.039230e-06 5.196152e-07
#>  [2,] 1.064890e-06 5.324452e-07
#>  [3,] 1.090551e-06 5.452753e-07
#>  [4,] 1.116211e-06 5.581053e-07
#>  [5,] 1.141871e-06 5.709353e-07
#>  [6,] 1.167531e-06 5.837653e-07
#>  [7,] 1.193191e-06 5.965953e-07
#>  [8,] 1.218851e-06 6.094253e-07
#>  [9,] 1.244511e-06 6.222553e-07
#> [10,] 1.270171e-06 6.350853e-07
#> 
#> $per_trait$traitX$tau_star_blocks_joint
#>          annot_A_0    annot_B_0
#>  [1,] 1.039230e-06 5.196152e-07
#>  [2,] 1.064890e-06 5.324452e-07
#>  [3,] 1.090551e-06 5.452753e-07
#>  [4,] 1.116211e-06 5.581053e-07
#>  [5,] 1.141871e-06 5.709353e-07
#>  [6,] 1.167531e-06 5.837653e-07
#>  [7,] 1.193191e-06 5.965953e-07
#>  [8,] 1.218851e-06 6.094253e-07
#>  [9,] 1.244511e-06 6.222553e-07
#> [10,] 1.270171e-06 6.350853e-07
#> 
#> $per_trait$traitX$h2g
#> [1] 0.3
#> 
#> $per_trait$traitX$nBlocks
#> [1] 10
#> 
#> 
#> $per_trait$traitY
#> $per_trait$traitY$summary
#> # A tibble: 2 × 20
#>   target    isBinary tauSingle tauSeSingle tauStarSingle tauStarSeSingle
#>   <chr>     <lgl>        <dbl>       <dbl>         <dbl>           <dbl>
#> 1 annot_A_0 TRUE     0.0000001  0.00000003   0.00000115      0.000000221
#> 2 annot_B_0 FALSE    0.0000001  0.00000003   0.000000577     0.000000111
#> # ℹ 14 more variables: enrichmentSingle <dbl>, enrichmentSeSingle <dbl>,
#> #   enrichmentPSingle <dbl>, enrichstatSingle <dbl>, enrichstatSeSingle <dbl>,
#> #   tauJoint <dbl>, tauSeJoint <dbl>, tauStarJoint <dbl>, tauStarSeJoint <dbl>,
#> #   enrichmentJoint <dbl>, enrichmentSeJoint <dbl>, enrichmentPJoint <dbl>,
#> #   enrichstatJoint <dbl>, enrichstatSeJoint <dbl>
#> 
#> $per_trait$traitY$tau_star_blocks_single
#>          annot_A_0    annot_B_0
#>  [1,] 1.039230e-06 5.196152e-07
#>  [2,] 1.064890e-06 5.324452e-07
#>  [3,] 1.090551e-06 5.452753e-07
#>  [4,] 1.116211e-06 5.581053e-07
#>  [5,] 1.141871e-06 5.709353e-07
#>  [6,] 1.167531e-06 5.837653e-07
#>  [7,] 1.193191e-06 5.965953e-07
#>  [8,] 1.218851e-06 6.094253e-07
#>  [9,] 1.244511e-06 6.222553e-07
#> [10,] 1.270171e-06 6.350853e-07
#> 
#> $per_trait$traitY$tau_star_blocks_joint
#>          annot_A_0    annot_B_0
#>  [1,] 1.039230e-06 5.196152e-07
#>  [2,] 1.064890e-06 5.324452e-07
#>  [3,] 1.090551e-06 5.452753e-07
#>  [4,] 1.116211e-06 5.581053e-07
#>  [5,] 1.141871e-06 5.709353e-07
#>  [6,] 1.167531e-06 5.837653e-07
#>  [7,] 1.193191e-06 5.965953e-07
#>  [8,] 1.218851e-06 6.094253e-07
#>  [9,] 1.244511e-06 6.222553e-07
#> [10,] 1.270171e-06 6.350853e-07
#> 
#> $per_trait$traitY$h2g
#> [1] 0.3
#> 
#> $per_trait$traitY$nBlocks
#> [1] 10
#> 
#> 
#> 
#> $meta
#> $meta$tauStar
#> # A tibble: 2 × 9
#>   target    isBinary  singleMean     singleSe  singleP nTraits jointMean jointSe
#>   <chr>     <lgl>          <dbl>        <dbl>    <dbl>   <int>     <dbl>   <dbl>
#> 1 annot_A_0 TRUE     0.00000115  0.000000156  1.52e-13       2   1.15e-6 1.56e-7
#> 2 annot_B_0 FALSE    0.000000577 0.0000000782 1.52e-13       2   5.77e-7 7.82e-8
#> # ℹ 1 more variable: jointP <dbl>
#> 
#> $meta$enrichment
#> # A tibble: 2 × 6
#>   target    isBinary singleMean singleSe  singleP nTraits
#>   <chr>     <lgl>         <dbl>    <dbl>    <dbl>   <int>
#> 1 annot_A_0 TRUE              2    0.283 0.000270       2
#> 2 annot_B_0 FALSE             2    0.283 0.000270       2
#> 
#> $meta$enrichstat
#> # A tibble: 2 × 6
#>   target    isBinary singleMean singleSe  singleP nTraits
#>   <chr>     <lgl>         <dbl>    <dbl>    <dbl>   <int>
#> 1 annot_A_0 TRUE         0.0556   0.0153 0.000270       2
#> 2 annot_B_0 FALSE        0.0556   0.0153 0.000270       2
#> 
#> 
#> $params
#> $params$maf_cutoff
#> [1] 0.05
#> 
#> $params$M_ref
#> [1] 6
#> 
#> $params$target_categories
#> [1] "annot_A_0" "annot_B_0"
#> 
#> $params$n_baseline
#> [1] 1
#> 
#> $params$baseline_categories
#> [1] "baselineLD_0"
#> 
#> $params$trait_names
#> [1] "traitX" "traitY"
#> 
#>

Usage

Arguments

Value

See also

Examples