Skip to contents

Fine-Map Epistatic SNP Pairs Within a Single Block

Source: R/epistasis.R
fine_map_epistasis_block.Rd

Exhaustively or adaptively identifies the specific SNP pairs within a single LD block that drive the block-level epistatic signal. Three methods are available:

"pairwise"

Exhaustive C(p,2) scan. Tests all pairs using the same model as scan_block_epistasis. Recommended for blocks with p <= 200 SNPs.

"lasso"

Fits a LASSO model with main effects and all pairwise interaction terms using glmnet. Identifies the interaction terms with non-zero coefficients at the cross-validated lambda. Recommended for blocks with p > 200 SNPs.

"auto"

Dispatches to "pairwise" when p <= 200, "lasso" otherwise.

Usage

fine_map_epistasis_block(
  block_id,
  geno_matrix,
  snp_info,
  blocks,
  y_resid,
  method = c("auto", "pairwise", "lasso"),
  min_freq = 0.05,
  sig_threshold = 0.05,
  sig_metric = c("p_simplem_sidak", "p_simplem", "p_bonf", "p_fdr"),
  meff_percent_cut = 0.995,
  lasso_nfolds = 5L,
  verbose = TRUE
)

Arguments

block_id

Character. The block to fine-map.

geno_matrix

Numeric matrix (individuals x SNPs) or LDxBlocks_backend.

snp_info

Data frame with columns SNP, CHR, POS.

blocks

LD block table.

y_resid

Named numeric vector of GRM-corrected REML residuals (from test_block_haplotypes null model). Names must match individual IDs.

method

Character. One of "pairwise", "lasso", "auto". Default "auto".

min_freq

Numeric. Minimum MAF. Default 0.05.

sig_threshold

Numeric. Significance threshold applied to the p-value chosen by sig_metric. Default 0.05.

sig_metric

Character. Which p-value drives the significant flag in pairwise output. One of "p_simplem_sidak" (default, recommended), "p_simplem", "p_bonf", or "p_fdr". All four p-value columns are always present regardless of this choice. Ignored for the "lasso" method (which returns lasso_coef and selected instead of p-values).

meff_percent_cut

Numeric in (0, 1). Variance threshold for simpleM eigendecomposition. Default 0.995.

lasso_nfolds

Integer. CV folds for glmnet lambda selection. Default 5L.

verbose

Logical. Default TRUE.

Value

A data frame sorted by p_wald (pairwise) or |coefficient| (lasso). Columns: SNP_i, SNP_j, POS_i, POS_j, dist_bp, aa_effect or lasso_coef, SE, t_stat, p_wald, p_bonf, significant (pairwise) or selected (lasso).

See also

scan_block_epistasis, scan_block_by_block_epistasis