triple_viz: Data visualization using PCA, t-SNE and UMAP

In vittoriofortino84/COPS: Clustering algorithms for Omics-driven Patient Stratification

Data visualization using PCA, t-SNE and UMAP

Description

Generates scatter plots from 2D embeddings.

Usage

triple_viz(
  data,
  category,
  category_label,
  tsne_perplexity = 45,
  umap_neighbors = 20,
  tsne = FALSE,
  pre_manifold_pca = TRUE,
  color_scale = scale_color_brewer(palette = "Dark2")
)

pca_viz(
  data,
  category,
  category_label,
  color_scale = scale_color_brewer(palette = "Dark2")
)

umap_viz(
  data,
  category,
  category_label,
  umap_neighbors = 20,
  pre_manifold_pca = TRUE,
  max_pcs = 50,
  color_scale = scale_color_brewer(palette = "Dark2")
)

tsne_viz(
  data,
  category,
  category_label,
  tsne_perplexity = 45,
  pre_manifold_pca = TRUE,
  color_scale = scale_color_brewer(palette = "Dark2")
)

Arguments

data	matrix with samples on rows
category	factor for coloring
category_label	name of color legend
tsne_perplexity	t-SNE perplexity parameter
umap_neighbors	UMAP neighbours parameter
tsne	whether to use t-SNE
pre_manifold_pca	whether to apply PCA before manifold learning (recommended for high-dimensional data)
color_scale	color scale used for category
max_pcs	maximum number of PCs (limited by data) to extract for UMAP

Value

list of plots

ggplot object

ggplot object

ggplot object

Functions

• pca_viz(): Data visualization using PCA

• umap_viz(): Data visualization using UMAP

• tsne_viz(): Data visualization using t-SNE

vittoriofortino84/COPS documentation built on Jan. 28, 2025, 3:16 p.m.