火山图

在分析RNAseq和microarray数据的差异基因的时候，常常用到火山图，需要的数据是包含Fold change（FC) 和pvalue的矩阵，通常横坐标用log2（FC）表示，
纵坐标用-log10（pvalue）表示。差异越大的基因位于左上角和右上角。

library(ggplot2)

#cell cycle pos
# RNAseq_genes 为FC和Pvalue的矩阵
# rownames(RNAseq_genes)<-RNAseq_genes$Row.names
#tiff("volcanoplot_cellcycle_pos.tiff",width = 550,height = 350)
RNAseq_genes_CC_pos<-RNAseq_genes[intersect(pos_cellcycle,rownames(RNAseq_genes)),]

image.png

RNAseq_genes_CC_pos$change <-  as.factor(ifelse(RNAseq_genes_CC_pos$PValue < 0.05 & abs(RNAseq_genes_CC_pos$logFC) > log2(1.5),ifelse(RNAseq_genes_CC_pos$logFC> log2(1.5),'up_genes','down_genes'),'not_DEgenes'))

ggplot(data = RNAseq_genes_CC_pos, aes(x = logFC, y = -log10(PValue), color = change)) +geom_point(alpha=0.8, size = 1) +
  theme_bw(base_size = 15) +
  theme(
    panel.grid.minor = element_blank(),
    panel.grid.major = element_blank()
  ) +scale_color_manual(name = "", values = c("red","blue", "grey"), limits = c('up_genes','down_genes','not_DEgenes')))

#dev.off()

image.png

加上显著基因的标签

library(ggrepel)
RNAseq_genes_CC_pos$sign <- ifelse(RNAseq_genes_CC_pos$PValue < 0.05 & abs(RNAseq_genes_CC_pos$logFC) > log2(2),rownames(RNAseq_genes_CC_pos),NA)

ggplot(data = RNAseq_genes_CC_pos, aes(x = logFC, y = -log10(PValue), color = change)) +geom_point(alpha=0.8, size = 1) +
  theme_bw(base_size = 15) +
  theme(
    panel.grid.minor = element_blank(),
    panel.grid.major = element_blank()
  ) +scale_color_manual(name = "", values = c("red","blue", "grey"), limits = c('up_genes','down_genes','not_DEgenes'))+geom_text_repel(aes(label =sign), size = 4.5,color = "black")

image.png

欢迎关注微信公众号：生信编程日常

公众号二维码.jpg

参考：
https://zhuanlan.zhihu.com/p/34992159
https://www.jianshu.com/p/003bdf9e6169