In GreshamLab/flowtools: Tools for Analyzing FlowCytometry Data

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

library(flowtools)

The goal of this vignette is to provide examples of commonly used workflows in the Gresham lab for performing flow cytometry analysis. It integrates existing tools from R packages that are used to analyze flow data as well as custom functions that are part of flowtools an R package written by the Gresham lab to simplify flow analysis.

The primary examples are motivated by copy number variant detetion using a CNV reporter, but the tools and methods are also applicable to other applications such as determining ploidy from FACS data or quantifying gene expression.

The code for this vignette and the flowtools package is available in the github repo.

Load Required libraries

Flow cytometry anlaysis in R requires several distinct libraries in addition to flowtools. The first step in any analysis is to load all libraries. In some cases, these libraries may first have to be downloaded using install.packages.

#install.packages(flowCore)
#install.packages(flowViz)
#install.packages(ggcyto)
#install.packages(ggforce)
#install.packages(tidyverse)
#install.packages(ggridges)
library(flowCore)
library(flowViz)
library(ggcyto)
library(ggforce)
library(tidyverse)
library(ggridges)
library(readxl)

Reading in Data

FCS files exported from different flow cytometry machines can be read into R To read in data we use read.ncdfFlowSet

###David's improved code below
dir = '~/Projects/data/flowdata/Accuri'
files <- sort(list.files(path=dir,pattern = ".fcs", full.names=TRUE))
flowData <- read.ncdfFlowSet(files=files, pattern=".fcs", alter.names = TRUE)

#needs to differ depending on sample sheet file type
sample.sheet <- read_excel(paste(path=dir,"GAP1_multicheck_20May2019.xlsx", sep="/"))


print(flowData)

Accuri Data

The raw Accuri data can be downloaded from here

#working directory
dir = '.'

#file location
#Used for data files being outside of home directory
path.data = '~/Projects/data/flowdata/Accuri'

#set name of run to create gates for
#Should be the same name as the folder that your .fcs files are in
name <- "Accuri"

#load sample sheet
sample.sheet <- read.csv(paste(path.data,"/samplesheet_",name,".csv", sep=""))

#read in fcs files in order presented in sample sheet (based on well identifier)
#files <- paste(path.data,"/",sort(factor(list.files(paste(path.data,"/", sep=""),full.names=FALSE), levels = paste(sample.sheet$Well,".fcs",sep="" ), ordered=TRUE)),sep="")

files <- sort(list.files(path=path.data,pattern = ".fcs", full.names=TRUE))
#files1 <- paste0(path.data, files, sep)
flowData <- read.ncdfFlowSet(files=files, pattern=".fcs", alter.names = TRUE)

#Ensures that the number of entries in the sample sheet match the number of flowFrames in the flowSet
sample.ind <- which(paste(sample.sheet$Well,".fcs", sep="") %in% sampleNames(flowData))
sample.sheet <- sample.sheet[sample.ind,]
sample.sheet <- sample.sheet[order(sample.sheet$Well),]

#rename sample name of flow set to make it easier to identify
sampleNames(flowData) <- paste(gsub(" ","_",sample.sheet$Strain),"_",sub(" ","_",sample.sheet$Well), sep="")