Fetches genotype data from a BrAPI server and converts it into a numeric
dosage matrix (samples × markers) compatible with genomic selection
packages like rrBLUP, BGLR, and sommer.
Arguments
- con
A
brapi_connection()object.- variantSetDbId
Character. The variant set to retrieve.
- sep
Character. Unphased allele separator. Default
"/"(e.g."0/1"). Phased calls using"|"are also handled automatically.- unknown_string
Character. String representing missing data. Default
".".
Value
A numeric matrix: rows = samples (callSetDbIds), columns = markers
(variantDbIds). Values are integer dosages (0, 1, 2 for diploids; 0–N
for polyploids). Missing calls are NA.
Details
Allele dosage is computed by splitting each genotype string on sep (or
"|" for phased calls) and counting how many alleles are non-reference
(i.e. not "0"). Missing calls (unknown_string, ".", or "") become
NA.
Examples
# \donttest{
con <- brapi_connection("https://test-server.brapi.org")
dosage <- brapi_get_dosage_matrix(con, "variantset1")
#> ℹ Fetching allele matrix for variant set "variantset1"...
#> ℹ Encoding 260 genotype calls as allele dosages...
#> ✔ Dosage matrix ready: 13 samples x 20 markers.
dim(dosage)
#> [1] 13 20
# Use with rrBLUP:
# library(rrBLUP)
# result <- mixed.solve(y = pheno$yield, Z = dosage)
# }