差异基因散点图绘制教程

本期教程

小杜的生信笔记，自2021年11月开始做的知识分享，主要内容是R语言绘图教程、转录组上游分析、转录组下游分析等内容。凡事在社群同学，可免费获得自2021年11月份至今全部教程，教程配备事例数据和相关代码，我们会持续更新中。

往期教程部分内容

绘图教程

导入所需R包

if (!require('ggplot2')) install.packages('ggplot2'); library('ggplot2')
if (!require('DESeq2')) install.packages('DESeq2'); library('DESeq2')
if (!require('biomaRt')) install.packages('biomaRt'); library('biomaRt')
if (!require('GEOquery')) install.packages('GEOquery'); library('GEOquery')
if (!require('vsn')) install.packages('vsn'); library('vsn')
if (!require('pheatmap')) install.packages('pheatmap'); library('pheatmap')

导入数据

cts <- read.csv("Input.data.csv",header = T, row.names = 1)

差异分析

##'@分组
sample_type = gsub(".*P7ma","Quiescent",gsub(".*P6ma","Cycling",colnames(cts)));
coldata = data.frame(row.names=colnames(cts), phenotype = sample_type ); 

#'@DESeq2差异分析
ddsMF <- DESeqDataSetFromMatrix(countData = cts, colData = coldata, design = ~ phenotype)
ddsMF <- DESeq(ddsMF)
resultsNames(ddsMF) 

#'get normalized counts'@标准化数据
nds <- DESeqTransform( SummarizedExperiment(log2(counts(ddsMF, normalized=TRUE) + 1), colData=colData(ddsMF)) ) # this gives log2(norm_cts + 1)
vsd <- vst(ddsMF, blind=FALSE) # variance stabilization

创建data.frame数据矩阵

quiescent = rownames(coldata)[coldata$phenotype=="Quiescent"]
cycling = rownames(coldata)[coldata$phenotype=="Cycling"]
data_cts_mets_avg = data.frame("quiescent"=rowMeans(assay(vsd)[, quiescent],na.rm=T),
                               "cycling"=rowMeans(assay(vsd)[, cycling],na.rm=T))

绘图

##'@显著性差异基因
upCQ = resCQ[!is.na(resCQ$padj) & resCQ$padj < 0.1 & resCQ$log2FoldChange > 1,]
dwCQ = resCQ[!is.na(resCQ$padj) & resCQ$padj < 0.1 & resCQ$log2FoldChange < -1,]
dim(upCQ)
dim(dwCQ)

plot(data_cts_mets_avg$quiescent, data_cts_mets_avg$cycling, pch=20, 
     xlab="mRNA expression quiescent cells (Normalized counts)", 
     ylab="mRNA expression cycling cells  (Normalized counts)",
     cex.lab=1.3,col=rgb(0.6,0.6,0.6,0.5));
abline(0,1,lty=2,col="black");
points(data_cts_mets_avg[rownames(upCQ),"quiescent"], data_cts_mets_avg[rownames(upCQ),"cycling"],pch=20,col=colors()[c(35)])
points(data_cts_mets_avg[rownames(dwCQ),"quiescent"], data_cts_mets_avg[rownames(dwCQ),"cycling"],pch=20,col=colors()[c(125)])
legend("topleft",legend=c("Up in Cycling","Down in Cycling"),col=colors()[c(35,125)],pch=20,inset=0.01,bty="n",cex=1.2)