Summarise Haplotype Allele Inventory Per Candidate Parent

Produces a tidy long-format allele inventory for each candidate parent individual, reporting which haplotype alleles they carry at each LD block and in what dosage. This is the primary decision support table for haplotype stacking crosses: it reveals which parents carry complementary rare alleles at important blocks, which candidates are genomically redundant (same alleles at all blocks), and which blocks should be targeted in crossing schemes.

The output includes one row per (individual x block x allele) combination, including rows where the individual carries 0 copies of an allele. This ensures that all candidates can be compared on the same rows for any given block.

Usage

summarize_parent_haplotypes(
  haplotypes,
  candidate_ids = NULL,
  allele_effects = NULL,
  blocks = NULL,
  min_freq = 0.02,
  missing_string = "."
)

Arguments

haplotypes

Named list produced by extract_haplotypes. Must carry a block_info attribute. All blocks in the list are included in the inventory unless filtered by candidate_ids or min_freq.

candidate_ids

Character vector of individual IDs to include in the inventory. Must match names in the haplotype vectors (i.e. names(haplotypes[[1]])). NULL (default) includes all individuals present in haplotypes. Supply the id column from score_favorable_haplotypes (e.g. top 20 by rank) to focus the inventory on selection candidates.

allele_effects

Optional data frame of per-allele effects to join to the inventory. When supplied, the allele_effect column is populated by matching on block_id and allele. NULL (default) leaves allele_effect as NA throughout. Required columns when not NULL:

block_id (character) - matching names(haplotypes).
allele (character) - matching haplotype allele strings.
allele_effect (numeric) - effect value per allele.

Accepts the output of decompose_block_effects directly. For test_block_haplotypes output, filter to one trait and rename effect to allele_effect first.

blocks

LD block table from run_Big_LD_all_chr. Used only as an alternative source of block coordinate metadata if the block_info attribute of haplotypes is missing or incomplete. Default NULL (uses block_info attribute).

min_freq

Numeric in (0, 1). Minimum allele frequency in the full panel (all individuals in haplotypes, not just candidate_ids). Alleles below this threshold are excluded from the inventory entirely. Default 0.02. This prevents the inventory from being dominated by private alleles observed in only one or two individuals.

missing_string

Character. String used in haplotype vectors to indicate missing genotype data. Individuals with this string at a block are skipped for that block (no rows emitted). Default ".".

Value

Data frame in long format. One row per individual x block x allele combination, including rows where dosage = 0 (allele absent). Sorted ascending by id, CHR, start_bp, then descending by dosage (so carried alleles appear before absent ones for each individual-block combination). Contains 10 columns:

id: Character. Individual identifier.
block_id: Character. LD block identifier matching names(haplotypes).
CHR: Character. Chromosome label.
start_bp: Integer. Block start coordinate (base pairs).
end_bp: Integer. Block end coordinate (base pairs).
allele: Character. Haplotype allele string (dosage-coded concatenation of SNP dosages within the block, e.g. "012102").
dosage: Integer. Number of copies of this allele carried by the individual: 0 = absent; 1 = one copy (heterozygous for phased data, or present for unphased data); 2 = two copies (homozygous, phased data only). For unphased input (no "|" in strings), dosage is always 0 or 1 - the two chromosomes cannot be distinguished, so an individual matching this allele string exactly receives dosage 1 regardless of whether they are truly homozygous or heterozygous.
allele_freq: Numeric in [0, 1]. Population frequency of this allele in the full panel (all individuals, not just candidates). Computed as table(valid_haplotypes) / length(valid_haplotypes) before any candidate_ids filtering is applied.
allele_effect: Numeric or NA. Effect value from allele_effects if supplied and the allele-block combination is matched; NA otherwise.
is_rare: Logical. TRUE when allele_freq < 0.10. Convenience flag for filtering the inventory to rare alleles that may be private to specific parents and therefore valuable for introgression.

Examples

# \donttest{
data(ldx_geno, ldx_snp_info, ldx_blocks, ldx_blues, package = "LDxBlocks")
haps <- extract_haplotypes(ldx_geno, ldx_snp_info, ldx_blocks)
pred <- run_haplotype_prediction(ldx_geno, ldx_snp_info, ldx_blocks,
                                  blues   = setNames(ldx_blues$YLD,
                                                     ldx_blues$id),
                                  verbose = FALSE)
ae <- decompose_block_effects(haps, ldx_snp_info, ldx_blocks,
                               snp_effects = pred$snp_effects)
# Inventory for top 5 candidates
top5 <- rownames(ldx_geno)[1:5]
inv  <- summarize_parent_haplotypes(haps, candidate_ids = top5,
                                     allele_effects = ae)
# Show blocks where candidates carry different alleles
inv[inv$dosage > 0, c("id","block_id","allele","dosage","allele_effect")]
#>         id              block_id                         allele dosage
#> 10  ind001    block_1_1000_25027      2022222002222002220020220      1
#> 60  ind001   block_1_81064_99022           22000002220202000020      1
#> 102 ind001 block_1_155368_179371      0110110010000000001010100      1
#> 156 ind001    block_2_1000_30023 111122112222221222211122211112      1
#> 259 ind001 block_2_161515_180473           11211121201111111221      1
#> 305 ind001    block_3_1000_19068           02000022220222000202      1
#> 353 ind001   block_3_74532_93854           20220202222222222220      1
#> 399 ind001 block_3_149647_168376           11112111211111112011      1
#> 64  ind002   block_1_81064_99022           11122211120121121200      1
#> 165 ind002    block_2_1000_30023 111011101001101100012000110001      1
#> 215 ind002  block_2_86236_105290           11100211112111111102      1
#> 270 ind002 block_2_161515_180473           22222220202222002222      1
#> 317 ind002    block_3_1000_19068           12111122221222111112      1
#> 357 ind002   block_3_74532_93854           01010020101211111011      1
#> 412 ind002 block_3_149647_168376           22112010110011102121      1
#> 27  ind003    block_1_1000_25027      1011121012111011210020110      1
#> 76  ind003   block_1_81064_99022           12111102120112111110      1
#> 129 ind003 block_1_155368_179371      1221121121212121112101112      1
#> 171 ind003    block_2_1000_30023 000022002222220222200222220002      1
#> 230 ind003  block_2_86236_105290           22200222222222222002      1
#> 271 ind003 block_2_161515_180473           00020020020222200222      1
#> 321 ind003    block_3_1000_19068           01000011210111110101      1
#> 374 ind003   block_3_74532_93854           21211212222112212210      1
#> 422 ind003 block_3_149647_168376           22202000220022202020      1
#> 33  ind004    block_1_1000_25027      0020000000000000002202000      1
#> 85  ind004   block_1_81064_99022           11122221211221121111      1
#> 138 ind004 block_1_155368_179371      1210120110111011112101111      1
#> 184 ind004    block_2_1000_30023 100012001111111111101111120001      1
#> 235 ind004  block_2_86236_105290           11100211112111111102      1
#> 283 ind004 block_2_161515_180473           00110021110111210221      1
#> 328 ind004    block_3_1000_19068           00011100101000210000      1
#> 381 ind004   block_3_74532_93854           11120111212222222121      1
#> 425 ind004 block_3_149647_168376           11022121101100012112      1
#> 50  ind005    block_1_1000_25027      2022222002222002220020220      1
#> 92  ind005   block_1_81064_99022           02222212111122222111      1
#> 142 ind005 block_1_155368_179371      0110110010000000001010100      1
#> 195 ind005    block_2_1000_30023 111011101001101100012000110001      1
#> 250 ind005  block_2_86236_105290           22200222222222222002      1
#> 293 ind005 block_2_161515_180473           00110021110111210221      1
#> 339 ind005    block_3_1000_19068           01000011210111110101      1
#> 394 ind005   block_3_74532_93854           21211212222112212210      1
#> 439 ind005 block_3_149647_168376           22112010110011102121      1
#>     allele_effect
#> 10      -0.020711
#> 60       0.042963
#> 102      0.060325
#> 156      1.265583
#> 259     -0.237346
#> 305     -0.543943
#> 353     -0.195629
#> 399      0.000421
#> 64       0.108826
#> 165      0.424797
#> 215     -0.298645
#> 270     -0.302974
#> 317     -0.068503
#> 357     -0.057814
#> 412      0.021767
#> 27      -0.054256
#> 76       0.125705
#> 129      0.601534
#> 171      1.043313
#> 230     -0.564878
#> 271     -0.512270
#> 321     -0.159478
#> 374     -0.106051
#> 422      0.059172
#> 33      -0.127989
#> 85       0.279913
#> 138      0.386147
#> 184      0.365229
#> 235     -0.298645
#> 283     -0.341994
#> 328      0.333311
#> 381     -0.153709
#> 425     -0.036984
#> 50      -0.020711
#> 92       0.379535
#> 142      0.060325
#> 195      0.424797
#> 250     -0.564878
#> 293     -0.341994
#> 339     -0.159478
#> 394     -0.106051
#> 439      0.021767
# }

Usage

Arguments

Value

See also

Examples