4.3 实战三 | 10X | 未过滤的PBMC
刘小泽写于2020.7.19
1 前言
相信学习单细胞数据分析的大家对PBMC都不陌生，虽然不是相关背景，但Seurat的PBMC数据深入人心。PBMC全称是peripheral blood mononuclear cell ，外周血单核细胞。
我们这里用的数据集来自Zheng et al. 2017，并在10X Genomics官网也可以获取，数量比Seurat使用的的2700个细胞数据更大
2 数据准备
下载
当然也可以跳过这一步，自己下载好之后读入。这里只是学习一下另一种方法
library(BiocFileCache)
bfc <- BiocFileCache("raw_data", ask = FALSE)
raw.path <- bfcrpath(bfc, file.path("http://cf.10xgenomics.com/samples",
"cell-exp/2.1.0/pbmc4k/pbmc4k_raw_gene_bc_matrices.tar.gz"))
untar(raw.path, exdir=file.path(tempdir(), "pbmc4k"))
# 最后也就是得到这三个文件：barcodes.tsv genes.tsv matrix.mtx读取
library(DropletUtils)
fname <- file.path(tempdir(), "pbmc4k/raw_gene_bc_matrices/GRCh38")
sce.pbmc <- read10xCounts(fname, col.names=TRUE)
# 看这里的数量惊人，但是后面还需要过滤
sce.pbmc
# class: SingleCellExperiment
# dim: 33694 737280
# metadata(1): Samples
# assays(1): counts
# rownames(33694): ENSG00000243485
# ENSG00000237613 ... ENSG00000277475
# ENSG00000268674
# rowData names(2): ID Symbol
# colnames(737280): AAACCTGAGAAACCAT-1
# AAACCTGAGAAACCGC-1 ... TTTGTCATCTTTAGTC-1
# TTTGTCATCTTTCCTC-1
# colData names(2): Sample Barcode
# reducedDimNames(0):
# altExpNames(0):转换ID，添加染色体信息
library(scater)
rownames(sce.pbmc) <- uniquifyFeatureNames(
rowData(sce.pbmc)$ID, rowData(sce.pbmc)$Symbol)
library(EnsDb.Hsapiens.v86)
location <- mapIds(EnsDb.Hsapiens.v86, keys=rowData(sce.pbmc)$ID,
column="SEQNAME", keytype="GENEID")3 质控
10X数据面临的一大问题就是空液滴，因此需要emptyDrops检验一下
set.seed(100)
# 对70多万个液滴进行检验
e.out <- emptyDrops(counts(sce.pbmc))
# > e.out
# DataFrame with 737280 rows and 5 columns
# Total LogProb PValue Limited FDR
# <integer> <numeric> <numeric> <logical> <numeric>
# AAACCTGAGAAACCAT-1 1 NA NA NA NA
# AAACCTGAGAAACCGC-1 0 NA NA NA NA
# AAACCTGAGAAACCTA-1 1 NA NA NA NA
# AAACCTGAGAAACGAG-1 0 NA NA NA NA
# AAACCTGAGAAACGCC-1 1 NA NA NA NA
# ... ... ... ... ... ...
# TTTGTCATCTTTACAC-1 2 NA NA NA NA
# TTTGTCATCTTTACGT-1 33 NA NA NA NA
# TTTGTCATCTTTAGGG-1 0 NA NA NA NA
# TTTGTCATCTTTAGTC-1 0 NA NA NA NA
# TTTGTCATCTTTCCTC-1 1 NA NA NA NA
# 最后过滤，剩下4000多
sce.pbmc <- sce.pbmc[,which(e.out$FDR <= 0.001)]
sce.pbmc
# class: SingleCellExperiment
# dim: 33694 4233
# metadata(1): Samples
# assays(1): counts
# rownames(33694): RP11-34P13.3 FAM138A ...
# AC213203.1 FAM231B
# rowData names(2): ID Symbol
# colnames(4233): AAACCTGAGAAGGCCT-1
# AAACCTGAGACAGACC-1 ... TTTGTCACAGGTCCAC-1
# TTTGTCATCCCAAGAT-1
# colData names(2): Sample Barcode
# reducedDimNames(0):
# altExpNames(0):数据备份
把unfiltered数据主要用在质控的探索上
unfiltered <- sce.pbmc这里过滤的条件不需要太严苛，只需要去除线粒体含量太高的细胞即可
成熟的mRNA会通过核孔来到细胞质。如果细胞遭到破坏，大量细胞质中mRNA流失，而线粒体体积比较大，流不出去，含量基本不变，最后导致细胞质中捕获的线粒体RNA占比升高
stats <- perCellQCMetrics(sce.pbmc, subsets=list(Mito=which(location=="MT")))
high.mito <- isOutlier(stats$subsets_Mito_percent, type="higher")
sce.pbmc <- sce.pbmc[,!high.mito]
summary(high.mito)
## Mode FALSE TRUE
## logical 3922 311根据原来的数据，加上质控标准作图
colData(unfiltered) <- cbind(colData(unfiltered), stats)
unfiltered$discard <- high.mito
gridExtra::grid.arrange(
plotColData(unfiltered, y="sum", colour_by="discard") +
scale_y_log10() + ggtitle("Total count"),
plotColData(unfiltered, y="detected", colour_by="discard") +
scale_y_log10() + ggtitle("Detected features"),
plotColData(unfiltered, y="subsets_Mito_percent",
colour_by="discard") + ggtitle("Mito percent"),
ncol=3
)再看下文库大小分别和线粒体含量的关系
plotColData(unfiltered, x="sum", y="subsets_Mito_percent",
colour_by="discard") + scale_x_log10()4 归一化
也是使用预分群+去卷积计算size factor的方法
library(scran)
set.seed(1000)
clusters <- quickCluster(sce.pbmc)
sce.pbmc <- computeSumFactors(sce.pbmc, cluster=clusters)
sce.pbmc <- logNormCounts(sce.pbmc)
summary(sizeFactors(sce.pbmc))
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.009 0.710 0.871 1.000 1.094 13.948看看两种归一化方法的差异
plot(librarySizeFactors(sce.pbmc), sizeFactors(sce.pbmc), pch=16,
xlab="Library size factors", ylab="Deconvolution factors", log="xy")
abline(a=0, b=1, col="red")5 找表达量高变化基因
既然有UMI，就可以用第三种方法【在之前 3.3 挑选高变化基因 的 2.3 考虑技术噪音中介绍过，如果没有spike-in】：可以考虑利用数据分布来表示技术噪音，例如只考虑技术噪音的话，UMI counts通常会呈现近似泊松分布
set.seed(1001)
dec.pbmc <- modelGeneVarByPoisson(sce.pbmc)
top.pbmc <- getTopHVGs(dec.pbmc, prop=0.1)
# 作图
plot(dec.pbmc$mean, dec.pbmc$total, pch=16, cex=0.5,
xlab="Mean of log-expression", ylab="Variance of log-expression")
curfit <- metadata(dec.pbmc)
# 可视化一条线（下图的蓝线），这条线指所有的基因都会存在的一种偏差
curve(curfit$trend(x), col='dodgerblue', add=TRUE, lwd=2)6 降维
set.seed(10000)
sce.pbmc <- denoisePCA(sce.pbmc, subset.row=top.pbmc, technical=dec.pbmc)
set.seed(100000)
sce.pbmc <- runTSNE(sce.pbmc, dimred="PCA")
set.seed(1000000)
sce.pbmc <- runUMAP(sce.pbmc, dimred="PCA")看看保留了几个PC
ncol(reducedDim(sce.pbmc, "PCA"))
## [1] 87 聚类
g <- buildSNNGraph(sce.pbmc, k=10, use.dimred = 'PCA')
clust <- igraph::cluster_walktrap(g)$membership
colLabels(sce.pbmc) <- factor(clust)
table(colLabels(sce.pbmc))
##
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
## 585 518 364 458 170 791 295 107 45 46 152 84 40 60 142 16 28 21
plotTSNE(sce.pbmc, colour_by="label")8 找marker基因并解释结果
markers <- findMarkers(sce.pbmc, pval.type="some", direction="up")这次看看cluster 7的marker基因
marker.set <- markers[["7"]]
as.data.frame(marker.set[1:30,1:3])
# p.value FDR summary.logFC
# FCN1 4.881588e-137 1.644802e-132 2.715872
# LGALS2 3.729029e-133 6.282295e-129 2.191398
# CSTA 1.426854e-131 1.602548e-127 2.123738
# CFD 1.207067e-102 1.016773e-98 1.503274
# FGL2 8.567117e-93 5.773209e-89 1.358891
# IFI30 7.822561e-80 4.392889e-76 1.276366
# CLEC7A 6.052032e-79 2.913102e-75 1.109366
# MS4A6A 1.958033e-78 8.246744e-75 1.419465
# CFP 8.802407e-73 3.295426e-69 1.312191
# S100A8 6.193215e-70 2.086742e-66 3.431603再和其他clusters对比
plotExpression(sce.pbmc, features=c("CD14", "CD68",
"MNDA", "FCGR3A"), x="label", colour_by="label")其实最后还是考验的背景知识：根据cluster7中CD14、CD68、MNDA表达量升高，同时又检查了CD16基因下调，推测cluster7是单核细胞
最后更新于
