README.md

In vallotlab/IDclust: Iterative Differential Clustering for single-cell

IDclust

IDclust is an unsupervised method for clustering single-cell datasets (scRNA, scEpigenome). The cells are iteratively re-processed and re-clustered. Subclusters are created provided they are significantly different from each other.

Installation

Install package

if (!require("devtools"))
    {
      install.packages("devtools", dep=TRUE)
        if(!require("devtools")) stop("Package not found")
    }

devtools::install_github("vallotlab/IDclust")

For scRNA use

install.packages("Seurat")

For scEpigenomics use

devtools::install_github("vallotlab/ChromSCape")

Usage

Running Iterative Differential Clustering

library(IDclust)
library(Seurat)

data("Seu") # for SingleCellExperiment - data("scExp")
Seu <- iterative_differential_clustering(
    Seu, # for SingleCellExperiment - scExp,
    output_dir = "./",
    plotting = TRUE,
    saving = TRUE,
    n_dims = 50,
    dim_red = "pca",
    vizualization_dim_red = "umap",
    logFC.th = log2(1.5),
    qval.th = 0.01
    )

Plotting network

plot_cluster_network(Seu) # for SingleCellExperiment - scExp

vallotlab/IDclust documentation built on Feb. 16, 2023, 8:58 a.m.