Skip to contents

Infer Structured Block-Level Diplotypes Per Individual

Source: R/haplotype_inference.R
infer_block_haplotypes.Rd

Converts the raw haplotype strings produced by extract_haplotypes into a structured per-individual, per-block diplotype table with explicit hap1 / hap2 gamete alleles, a diplotype code, a heterozygosity flag, and a phase-ambiguity flag.

When the input is a phased list (from read_phased_vcf or a pre-phased VCF via read_phased_vcf), hap1 and hap2 are the true gametic strings and phase_ambiguous is always FALSE.

When the input is unphased (diploid allele strings), heterozygous individuals have two possible phase assignments. The function flags these with phase_ambiguous = TRUE; hap1 / hap2 are set to NA for those individuals unless resolve_unphased = TRUE, in which case the most-frequent gametic split consistent with observed allele frequencies is imputed (maximum-parsimony heuristic - not statistically rigorous, use phase_with_beagle for rigorous phasing).

Usage

infer_block_haplotypes(
  haplotypes,
  resolve_unphased = FALSE,
  missing_string = "."
)

Arguments

haplotypes

Named list from extract_haplotypes. If produced from a phased list input, strings contain "|" separators (e.g. "011|100"). If unphased, strings are plain diploid allele codes (e.g. "012201").

resolve_unphased

Logical. For unphased heterozygous genotypes, impute the most parsimonious hap1/hap2 split using population allele frequencies (parsimony heuristic). Default FALSE (leave as NA).

missing_string

Character. Missing data placeholder. Default ".".

Value

Data frame with one row per individual x block combination:

block_id

Block identifier.

CHR, start_bp, end_bp

Block coordinates.

id

Individual ID.

hap1

Gamete 1 allele string (NA if unresolvable).

hap2

Gamete 2 allele string (NA if unresolvable).

diplotype

Canonical diplotype code: the two gamete alleles sorted alphabetically and joined with "/", e.g. "010/110".

heterozygous

Logical; TRUE when the individual is biologically heterozygous. For phased input: hap1 != hap2. For unphased input: any dosage position equals 1, regardless of whether phase was resolved.

phase_ambiguous

Logical; TRUE for unphased heterozygous individuals when resolve_unphased = FALSE. These individuals have heterozygous = TRUE but hap1 = hap2 = NA.

missing

Logical; TRUE when the allele string contains the missing placeholder.

See also

extract_haplotypes, phase_with_beagle, collapse_haplotypes

Examples

# \donttest{
data(ldx_geno, ldx_snp_info, ldx_blocks, package = "LDxBlocks")
haps <- extract_haplotypes(ldx_geno, ldx_snp_info, ldx_blocks)
dip  <- infer_block_haplotypes(haps)
# Proportion of heterozygous diplotypes per block
tapply(dip$heterozygous, dip$block_id, mean, na.rm = TRUE)
#>    block_1_1000_25027 block_1_155368_179371   block_1_81064_99022 
#>             0.7500000             0.7833333             0.6916667 
#>    block_2_1000_30023 block_2_161515_180473  block_2_86236_105290 
#>             0.7750000             0.7916667             0.7166667 
#>    block_3_1000_19068 block_3_149647_168376   block_3_74532_93854 
#>             0.7583333             0.8000000             0.7000000 
# }