如果传统bulk转录组数据队列足够大也可以使用单细胞流程

Posted on by

昨天我在生信菜鸟团分享的学徒数据挖掘任务: 不一定正确的多分组差异分析结果热图展现 提到了可以使用单细胞转录组数据分析流程来处理文献的数据集。

还给出了一些简单代码，就是看看样本聚类情况，然后留下作业给另外一个学徒，看单细胞R包Seurat的FindAllMarkers函数对7个亚型找到的marker基因，根据传统的bulk转录组差异分析策略的差异。

先看看单细胞转录组代码

这里我们的单细胞转录组数据分析方法，基本上遵循我的全网第一个单细胞课程（基础）满一千份销量就停止发售 内容，就是一些R包的认知，包括 scater,monocle,Seurat,scran,M3Drop 需要熟练掌握它们的对象，：一些单细胞转录组R包的对象 ，分析流程也大同小异：

  • step1: 创建对象
  • step2: 质量控制
  • step3: 表达量的标准化和归一化
  • step4: 去除干扰因素(多个样本整合)
  • step5: 判断重要的基因
  • step6: 多种降维算法
  • step7: 可视化降维结果
  • step8: 多种聚类算法
  • step9: 聚类后找每个细胞亚群的标志基因
  • step10: 继续分类

代码如下：

library(Seurat)
# https://satijalab.org/seurat/mca.html
sce <- CreateSeuratObject(counts = sarc_choose, 
 meta.data =clinic,
 min.cells = 5, min.features = 1000, 
 project = "sce")
head(sce@meta.data) 
# 步骤 ScaleData 的耗时取决于电脑系统配置（保守估计大于一分钟）
sce <- ScaleData(object = sce, 
 vars.to.regress = c('nCount_RNA'), 
 model.use = 'linear', 
 use.umi = FALSE)
sce <- FindVariableFeatures(object = sce, 
 mean.function = ExpMean, 
 dispersion.function = LogVMR, 
 x.low.cutoff = 0.0125, 
 x.high.cutoff = 4, 
 y.cutoff = 0.5)
length(VariableFeatures(sce)) 
sce <- RunPCA(object = sce, pc.genes = VariableFeatures(sce))
# 下面只是展现不同降维算法而已，并不要求都使用一次。
sce <- RunICA(sce )
sce <- RunTSNE(sce )
#sce <- RunUMAP(sce,dims = 1:10)
#VizPCA( sce, pcs.use = 1:2)
DimPlot(object = sce, reduction = "pca") 
DimPlot(object = sce, reduction = "ica")
DimPlot(object = sce, reduction = "tsne",group.by = 'short.histo')

# 针对PCA降维后的表达矩阵进行聚类 FindNeighbors+FindClusters 两个步骤。
sce <- FindNeighbors(object = sce, dims = 1:20, verbose = FALSE) 
sce <- FindClusters(object = sce, resolution = 0.8,verbose = FALSE)
# 继续tSNE可视化
DimPlot(object = sce, reduction = "tsne",
 group.by = 'RNA_snn_res.0.8',label = T)
table(sce@meta.data$short.histo,sce@meta.data$RNA_snn_res.0.8)

Idents(sce)=sce@meta.data$short.histo

sce.markers <- FindAllMarkers(object = sce, only.pos = TRUE, min.pct = 0.25, 
 thresh.use = 0.25)
DT::datatable(sce.markers)
library(dplyr)
sce.markers=sce.markers[!is.infinite(sce.markers$avg_logFC),]
sce.markers %>% group_by(cluster) %>% top_n(2, avg_logFC)
topN <- sce.markers %>% group_by(cluster) %>% top_n(50, avg_logFC)
DoHeatmap(sce,topN$gene)

热图如下：

image-20191008145324990

可以看到R包Seurat的FindAllMarkers函数对7个亚型找到的marker基因基本上都是上调基因。

检查单细胞转录组和bulk差异分析结果重合情况

首先bulk差异分析策略见: 不一定正确的多分组差异分析结果热图展现 ，其实就是我们以前在生信技能树分享的一个策略：如果你的分组比较多，差异分析策略有哪些？ 把其中一个亚型跟所有的其它样本进行常规转录组差异分析而已。差异分析结果如下：

image-20191008145724214

我们的单细胞转录组分析结果的，topN 这个变量，就存储着其差异分析结果：

image-20191008145634358

现在要比较这两个表格！

单细胞FindAllMarkers并不是简单取差异最大的基因

通常，我们对传统bulk转录组差异分析结果，可以选取top的上下调基因进行热图可视化，如下：

image-20191008160251314

但是，我们上面单细胞流程的R包Seurat的FindAllMarkers函数对ULMS亚型找到的marker基因，却并不满足这个传统bulk转录组差异分析统计学显著指标，比如logFC大于2，并且校正后的P值小于0.05

不过，也是可以看到这样的规律，就是挑选的基因可以显著的把绝大部分ULMS 亚型病人跟其他病人区分开来。

image-20191008160930446

但是，这些基因，在limma的voom算法的差异分析结果里面，只有非常少的几个基因的阈值是达标的：

image-20191008161055537

绝大部分基因的logFC机会没有啥意义。

其实FindAllMarkers找基因还有很多参数可以选择

有趣的是，就是没有limma的voom算法！

"wilcox" : Identifies differentially expressed genes between two groups of cells using a Wilcoxon Rank Sum test (default)
"bimod" : Likelihood-ratio test for single cell gene expression, (McDavid et al., Bioinformatics, 2013)
"roc" : Identifies 'markers' of gene expression using ROC analysis. For each gene, evaluates (using AUC) a classifier built on that gene alone, to classify between two groups of cells. An AUC value of 1 means that expression values for this gene alone can perfectly classify the two groupings (i.e. Each of the cells in cells.1 exhibit a higher level than each of the cells in cells.2). An AUC value of 0 also means there is perfect classification, but in the other direction. A value of 0.5 implies that the gene has no predictive power to classify the two groups. Returns a 'predictive power' (abs(AUC-0.5) * 2) ranked matrix of putative differentially expressed genes.
"t" : Identify differentially expressed genes between two groups of cells using the Student's t-test.
"negbinom" : Identifies differentially expressed genes between two groups of cells using a negative binomial generalized linear model. Use only for UMI-based datasets
"poisson" : Identifies differentially expressed genes between two groups of cells using a poisson generalized linear model. Use only for UMI-based datasets
"LR" : Uses a logistic regression framework to determine differentially expressed genes. Constructs a logistic regression model predicting group membership based on each feature individually and compares this to a null model with a likelihood ratio test.
"MAST" : Identifies differentially expressed genes between two groups of cells using a hurdle model tailored to scRNA-seq data. Utilizes the MAST package to run the DE testing.
"DESeq2" : Identifies differentially expressed genes between two groups of cells based on a model using DESeq2 which uses a negative binomial distribution (Love et al, Genome Biology, 2014).This test does not support pre-filtering of genes based on average difference (or percent detection rate) between cell groups. However, genes may be pre-filtered based on their minimum detection rate (min.pct) across both cell groups. To use this method, please install DESeq2, using the instructions at https://bioconductor.org/packages/release/bioc/html/DESeq2.html

感兴趣的朋友，可以一一尝试，当然了，还不如系统性学习下统计学知识哦

可是，统计学不比尝试这些参数要简单其实。

​

Comments are closed.