Contents
reduction
Tag a matrix as an ordered-basis embedding
Tags value so that study$samples$<name> <- value files it as an embedding rather than as a
plain table member. compute$populationStructure() and compute$embedding$samples() attach the
tag themselves, so the common path never calls reduction() directly.
Usage
reduction(value, method = NULL, source = NULL)Arguments
| Argument | Description |
|---|---|
value | A matrix or matrix-like object. Embedding coordinates, one row per sample, in the study's own sample order. |
method | Character scalar, or NULL (default). Reduction method, e.g. "pca" or "umap". |
source | Character scalar, or NULL (default). What the reduction was computed against -- a reference-panel id, or "self" for an in-sample computation. |
Value
value, unchanged apart from a reduction attribute carrying
list(method = method, source = source).
Examples
pcs <- matrix(rnorm(20), nrow = 10, dimnames = list(NULL, c("PC1", "PC2")))
attr(reduction(pcs, method = "pca", source = "1000G"), "reduction")See Also
Other study:
PolyGeniusFrame(),
PolyGeniusFrameSet(),
PolyGeniusStudy(),
align(),
roles()