Skip to contents

Locus novelty check using LD overlap

Source: R/novelty_check.R
novelty_check.Rd

This function assesses whether query loci are novel or match known loci by combining genomic proximity (± flanking window) and linkage disequilibrium (LD).

Usage

novelty_check(
  known_loci,
  query_loci,
  ldops = NULL,
  flanking = 1e+06,
  pop = "EUR",
  verbose = TRUE
)

Arguments

known_loci

Data.frame of known/published loci. Must contain columns: chr, pos, uniprot, rsid, prot.

query_loci

Data.frame of query loci to evaluate. Must contain columns: chr, pos, uniprot, rsid, prot.

ldops

Optional list specifying local LD computation:

bfile

PLINK binary prefix (bed/bim/fam)

plink

Path to PLINK executable

flanking

Genomic window (bp) around query loci used for overlap. Default is 1e6 (±1 Mb).

pop

1000 Genomes population code (e.g., "EUR") used when ldops = NULL.

verbose

Logical; if TRUE prints missing LD variants.

Value

A data.frame with paired known/query loci and LD r^2 values.

Details

It supports:

  • 1000 Genomes LD reference via ieugwasr

  • Local PLINK reference panels via ld_matrix_local()

The function:

  1. Finds overlapping loci within a flanking distance

  2. Matches loci by gene/protein (uniprot)

  3. Computes LD (r and r^2) between known and query variants

  4. Returns per-pair LD values for downstream novelty/replication assessment

Examples

if (FALSE) { # \dontrun{
# 1000G mode
novelty_check(known_loci, query_loci)

# Local PLINK mode
novelty_check(
  known_loci,
  query_loci,
  ldops = list(
    bfile = "/path/interval.imputed.olink.chr_3",
    plink = "/path/plink"
  )
)
} # }