This function performs clustering using the specified algorithm (DBSCAN or GMM) on UMAP or fPCA coordinates.
Usage
clustering(object, ...)
# S4 method for class 'DGrowthR'
clustering(
object,
embedding = "umap",
algorithm = "dbscan",
k = 0.01,
eps = NULL
)Arguments
- object
A DGrowthR object containing preprocessed data including UMAP coordinates.
- ...
Additional parameters to clustering algorithm
- embedding
A string indicating if the coordinates from "umap" or "fpca" should be used for clustering.
- algorithm
A string specifying the clustering algorithm to use ("dbscan" or "gmm").
- k
Numeric value indicating the number of nearest neighbors (for DBSCAN) or number of mixture components (for GMM). If k < 1, it is interpreted as a proportion of the total number of curves.
- eps
Numeric value indicating the epsilon distance to use for DBSCAN clustering. If NULL, automatic selection based on k is performed.