Write Marginal Analysis Results to GDS Annotation Node

Source: R/marginal_scan.R

Stores marginal analysis results as an annotation node in a GDS file for downstream reuse. Follows the same pattern as FunctionalAnnotation storage.

annotate_gds_marginal(
  gds_path,
  results,
  node_name = "MarginalAnalysis",
  overwrite = FALSE,
  verbose = 1
)

Arguments

gds_path

Character path to GDS file to annotate.

results

A data.frame (class "glow_marginal_scan" or plain data.frame) from marginal_scan().

node_name

Character name for the annotation node. Default: "MarginalAnalysis".

overwrite

Logical. If TRUE, replace existing node. Default: FALSE.

verbose

Integer verbosity level (0=silent, 1=progress).

Value

Invisible gds_path.

Details

Opens the GDS file for writing and navigates to (or creates) annotation/info/<node_name>. The results data.frame is aligned to the GDS variant order via match(gds_variant_ids, results$variant_id); any GDS variant not present in results will have NA values in the stored matrix. A verbose message is emitted when unmatched variants are detected.

Columns stored: MAF, MAC, score, var_score, Z, pvalue. When SPA results are present in results, Z_SPA and pvalue_SPA are also stored. The matrix is compressed with LZMA_RA. Metadata attributes "column_names" and "analysis_date" are attached to the node.

Complexity: O(V) in the total number of GDS variants. The file is modified in-place; no temporary copy is created.

Examples

if (FALSE) { # \dontrun{
pheno   <- extract_pheno_covar_gds("chr22.gds", covar_names = "sex")
nm      <- fit_null_model(pheno$X, pheno$Y, trait = "binary",
                          sample_id = pheno$sample_id)
results <- marginal_scan("chr22.gds", nm, use_SPA = FALSE)
annotate_gds_marginal("chr22.gds", results, node_name = "MarginalAnalysis")
} # }