README.md

In liuqivandy/scUnifrac: a package for quantitative assessment of cell population diversity in single-cell landscapes

scUnifrac

• Introduction
• Installation
• Usage
• Report example

Introduction

scUnifrac is a R package to quantify cell population diversity between two single-cell transcriptome profiles, which calculates the distance, estimates the statistical significance of the distance, identifies the subpopulations that are significant different between two profiles, finds genes that mark subpopulations, and predicts cell types of subpopulations. If two single-cell RNA-seq profiles have identical cell subpopulation structures (the same subpopulations with the same proportion), the distance is zero, whereas, the distance is one if completely different subpopulations. Besides calculating the distance and the pvalue, scUnifrac will generate a report , which include results summary and figures illustrating subpopulation structure, marker genes and predicted cell types in each subpopulation.

Citation

If you use scUnifrac, please cite the paper. Qi Liu, Charles A. Herring et al.,Quantitative assessment of cell population diversity in single-cell landscapes PLoS Biol. 2018 Oct; 16(10): e2006687.

Installation

To generate a report , you need to install pandoc (https://github.com/jgm/pandoc/releases/tag/2.2.1) . After installation , update your path to include the directory where pandoc’s binaries are installed.

To install scUnifrac, use

source("https://bioconductor.org/biocLite.R")
BiocManager::install("liuqivandy/scUnifrac")
#biocLite("liuqivandy/scUnifrac")

Usage

After installing scUnifrac, use following codes to run examples

library(scUnifrac)

#load two example datasets, one is from mouse colon, the other is from mouse pancreas
load(system.file("extdata", "colon1.Rdata", package = "scUnifrac"))
load(system.file("extdata", "pan1.Rdata", package = "scUnifrac"))

##run scUnifrac on the two example datasets without generating a report
result<-scUnifrac(data1=colon1,data2=pan1,report=F)
result

## when run scUnifrac, use a reference datasets to predict cell types and generate a report in the work directory
#load Mouse cell atlas [Han et al., 2018, Cell 172, 1091–1107]. The atlas is used as a reference to predict cell types
load(system.file("extdata", "ref.expr.Rdata", package = "scUnifrac"))
result<-scUnifrac(data1=colon1,data2=pan1,ref.expr=ref.expr,report=T)

#run scUnifrac on two similar samples
ind<-sample(c(1:ncol(colon1)), ncol(colon1)/2)
result<-scUnifrac(data1=colon1[,ind],data2=colon1[,-ind],report=F)

#run scUnifrac with the cluster information that user provides (e.g., cluster from other software)
ind<-sample(c(1:ncol(colon1)), ncol(colon1)/2)
clusterID<-sample(1:20, dim(colon1)[2],replace=T)
result<-scUnifrac(data1=colon1[,ind],data2=colon1[,-ind],cluster=clusterID, report=F)

#run scUnifrac_multi to calculate pairwise distance between multiple (>=2) scRNA-seq samples
#generate a simulated dataset including three samples, two from colon, one from pancreas
combineddata<-cbind(colon1[,1:500],pan1[,1:500],colon1[,501:1000])
group<-c(rep("colon1_1",500),rep("pan1",500),rep("colon1_2",500))
result<-scUnifrac_multi(dataall=combineddata,group=group)

#run scUnifrac_predictCelltype to predict cell types of query cells
load(system.file("extdata", "ref.expr.Rdata", package = "scUnifrac"))
preCell<-scUnifrac_predictCelltype(colon1[,1:500],ref.expr=ref.expr,normalize=T)
#only report cell types having correlation greater than 0.3 with at least one query cells
preCell<-scUnifrac_predictCelltype(colon1[,1:500],ref.expr=ref.expr,normalize=T,corcutoff=0.3)

liuqivandy/scUnifrac documentation built on Jan. 21, 2021, 2:02 p.m.