Retrieves a filtered subset of a pedigree tree via /pedigree - a batch
endpoint for pulling pedigree records across many germplasm in one call.
This is different from brapi_germplasm_pedigree(), which retrieves one
germplasm's immediate pedigree via the germplasm sub-resource
(/germplasm/{germplasmDbId}/pedigree) and must be called once per
germplasm. Use brapi_pedigree() (or brapi_search_pedigree()) to pull
pedigree records for many germplasm at once - e.g. everything in a crop,
program, or family - in one or a few requests; use
brapi_germplasm_pedigree() when you already have a single germplasm ID
in hand.
Usage
brapi_pedigree(
con,
germplasmDbId = NULL,
includeParents = NULL,
includeSiblings = NULL,
includeProgeny = NULL,
includeFullTree = NULL,
pedigreeDepth = NULL,
progenyDepth = NULL,
...
)Arguments
- con
A
brapi_connection()object.- germplasmDbId
Character or NULL. Filter by germplasm.
- includeParents
Logical or NULL. Include each node's parents.
- includeSiblings
Logical or NULL. Include each node's siblings.
- includeProgeny
Logical or NULL. Include each node's progeny.
- includeFullTree
Logical or NULL. Recursively include every node reachable in the pedigree tree.
- pedigreeDepth
Integer or NULL. Number of levels to include up the tree (parents, grandparents, ...).
- progenyDepth
Integer or NULL. Number of levels to include down the tree (children, grandchildren, ...).
- ...
Additional query parameters.
Value
A tibble with one row per pedigree node. parents, siblings,
and progeny, when requested, are list-columns of small tibbles (one
row per relative) rather than raw nested lists or a flattened table.
Details
Each row is one pedigree node (one germplasm). The server only includes
a node's relatives if asked: set includeParents, includeSiblings,
and/or includeProgeny to TRUE to populate the parents, siblings,
and progeny list-columns, each holding a small tibble of related
germplasm (germplasmDbId, germplasmName, and parentType - NA for
siblings, which have none) that you can tidyr::unnest() when you need
one row per relationship rather than one row per node. Nodes are never
collapsed or flattened by default: a pedigree is graph-shaped (each node
has its own parents, siblings, and progeny edges), and the three
relation types don't share a common row shape, so there is no lossless
single flat table to fall back to.
BrAPI endpoint
GET /pedigree - see the
v2.1 specification.
Query parameters the specification defines, which may be passed
through ...:
accessionNumber, collection, familyCode, binomialName, genus,
species, synonym, includeParents, includeSiblings,
includeProgeny, includeFullTree, pedigreeDepth, progenyDepth.
See also
brapi_germplasm_pedigree() for one germplasm's pedigree via
the germplasm sub-resource; brapi_search_pedigree() for the same
batch retrieval via POST, with a fuller set of filters.
Examples
# \donttest{
con <- brapi_connection("https://test-server.brapi.org")
brapi_pedigree(con, includeParents = TRUE, includeProgeny = TRUE)
#> # A tibble: 3 × 15
#> additionalInfo externalReferences breedingMethodDbId breedingMethodName
#> <lgl> <lgl> <chr> <chr>
#> 1 NA NA breeding_method1 Male Backcross
#> 2 NA NA breeding_method1 Male Backcross
#> 3 NA NA breeding_method1 Male Backcross
#> # ℹ 11 more variables: crossingProjectDbId <chr>, crossingYear <int>,
#> # defaultDisplayName <chr>, familyCode <chr>, germplasmDbId <chr>,
#> # germplasmName <chr>, germplasmPUI <chr>, parents <list>,
#> # pedigreeString <chr>, progeny <list>, siblings <list>
# }