UnsupClust_UMAP_proc: Perform unsupervised clustering and UMAP reduction
In eisascience/scCustFx: single-cell custom fxs Package

UnsupClust_UMAP_proc

R Documentation

Perform unsupervised clustering and UMAP reduction

Description

This function performs unsupervised clustering on single-cell RNA-seq data and then it does the Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction method. It returns and updated seurat object.

Usage

UnsupClust_UMAP_proc(
  SerObj,
  doClust = T,
  Feats = NULL,
  reduction = NULL,
  dims = NULL,
  reSerObjlution = 0.4,
  verbose = F,
  random.seed = 1234,
  doUMAP = T,
  n.components = 2,
  n.neighbors = 60,
  n.epochs = 300,
  min.dist = 0.2,
  spread = 1,
  reduction.key = "UMAPCust_"
)

Arguments

`SerObj`	A Seurat obj
`doClust`	Logical if F unsupervised clustering is only done default (T)
`Feats`	A vector of character strings specifying which features to use for clustering. If NULL, all features are used.
`reduction`	The type of dimensionality reduction to apply. If NULL, no reduction is applied. Options are "PCA", "TSNE", "UMAP", or "none".
`dims`	The number of dimensions to use for the reduction. If NULL, the default number of dimensions for the chosen method is used.
`reSerObjlution`	The UMAP reSerObjlution parameter.
`verbose`	A boolean indicating whether to print progress updates during the clustering process.
`random.seed`	The random seed to use for reproducibility.
`doUMAP`	Logical if F unsupervised clustering is only done default (T)
`n.components`	The number of umap components default 2, or 3 or try 1
`n.neighbors`	The number of neighbors to use for UMAP construction.
`n.epochs`	The number of epochs to use for UMAP construction.
`min.dist`	The minimum distance between UMAP points.
`spread`	The spread parameter for UMAP.
`reduction.key`	A character string specifying the prefix for the output object names.