Clustering is the key step to define cell types from the heterogeneous cells population. One critical challenge is that single-cell RNA sequencing (scRNA-Seq) experiments generate a massive amount of data with an excessive noise level, which hinders many computational algorithms. We propose a single cell Clustering method using Autoencoder and Network fusion (scCAN) that can accurately segregate cells from the high-dimensional noisy scRNA-Seq data.

Analysis on the sample dataset

#Load example data (SCE dataset)

#Get data matrix and label
data <- t(SCE$data); label <- as.character(SCE$cell_type1)


#Generate clustering result, the input matrix has rows as samples and columns as genes
result <- scCAN(data, r.seed = 1)

#The clustering result can be found here 
cluster <- result$cluster

#Calculate adjusted Rand Index using mclust package
ari <- round(scCAN::adjustedRandIndex(cluster,label), 2)
print(paste0("ARI = ", ari))