In zhiyhu/CIDER: Meta-Clustering for Single-Cell Data Integration and Evaluation

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

Introduction

This vignette performs dnCIDER on a cross-species pancreas dataset.

Set up

In addition to CIDER, we will load the following packages:

library(CIDER)
library(Seurat)
library(parallel)
library(cowplot)

Load pancreas data

The example data can be downloaded from https://figshare.com/s/d5474749ca8c711cc205.

Pancreatic cell data$^1$ contain cells from human (8241 cells) and mouse (1886 cells).

load("../data/pancreas_counts.RData") # count matrix
load("../data/pancreas_meta.RData") # meta data/cell information
seu <- CreateSeuratObject(counts = pancreas_counts, meta.data = pancreas_meta)
table(seu$Batch)

Perform dnCIDER (high-level)

DnCIDER contains three steps

seu <- initialClustering(seu, additional.vars.to.regress = "Sample", dims = 1:15)
ider <- getIDEr(seu, downsampling.size = 35, use.parallel = FALSE, verbose = FALSE)
seu <- finalClustering(seu, ider, cutree.h = 0.35) # final clustering

Visualise clustering results

We use the Seurat pipeline to perform normalisation (NormalizeData), preprocessing (FindVariableFeatures and ScaleData) and dimension reduction (RunPCA and RunTSNE).

seu <- NormalizeData(seu, verbose = FALSE)
seu <- FindVariableFeatures(seu, selection.method = "vst", nfeatures = 2000, verbose = FALSE)
seu <- ScaleData(seu, verbose = FALSE)
seu <- RunPCA(seu, npcs = 20, verbose = FALSE)
seu <- RunTSNE(seu, reduction = "pca", dims = 1:12)

We can see

scatterPlot(seu, "tsne", colour.by = "CIDER_cluster", title = "asCIDER clustering results") 

By comparing the dnCIDER results to the cell annotation from the publication$^1$, we observe that dnCIDER correctly identify the majority of populations across two species.

scatterPlot(seu, "tsne", colour.by = "Group", title = "Ground truth of cell populations") 

Technical

sessionInfo()

References

1. Baron, M. et al. A Single-Cell Transcriptomic Map of the Human and Mouse Pancreas Reveals Inter- and Intra-cell Population Structure. Cell Syst 3, 346–360.e4 (2016).
2. Satija R, et al. Spatial reconstruction of single-cell gene expression data. Nature Biotechnology 33, 495-502 (2015).

zhiyhu/CIDER documentation built on Feb. 4, 2025, 1:09 a.m.