#调用包及载入数据library(ggrepel)library(tidyverse)library(ggplot2)    #这个函数的确能读入数据，但是通过dim函数可以发现该文件具有31338行和35列，而35列并不是我们全部需要的，可以在载入数据的时候进行数据筛选#dt <- read.delim('ArFvsArP_deg_all.xls')#View(dt)

#dim(dt)

#筛选数据dt <- read.delim('C:\Users\10395\Desktop\r画图教程\数据\1.9火山图\ArFvsArP_deg_all.xls') %>%  select(GeneID,ArP_readcount,ArF_readcount,log2FoldChange,pval) %>%  filter(ArP_readcount > 0 | ArF_readcount > 0) %>%  column_to_rownames(var = 'GeneID')#更改列名colnames(dt) <- c('arp','arf','logFC','Pval')head(dt)

#添加catogrey行，并对每一个基因的状态进行分类    dt$catogrey <- if_else(dt$logFC > 2 & dt$Pval<0.001,'up',        if_else(dt$logFC < -2 & dt$Pval<0.001,'down','normal'))head(dt)

#获得catogrey的状态,以下是一样的效果dt %>% group_by(catogrey) %>% summarise(freq = n())#table(dt$catogrey)

#添加name行，为加标签方便dt$name <- rownames(dt)########开始进行火山图绘制# 简单火山图的绘制p1 <- ggplot(data = dt,             aes(x = logFC,                 y = -log10(Pval))) +  # 设置x轴为logFC，y轴为-P.Value的对数值  geom_point(alpha = 0.4, size = 2.5,                  aes(color = catogrey))           p1

#改变颜色p2 <- ggplot(data = dt,             aes(x = logFC,                 y = -log10(Pval))) +  # 设置x轴为logFC，y轴为-P.Value的对数值  geom_point(alpha = 0.4, size = 2.5,             aes(color = catogrey)) +      # 添加散点，根据change列着色  ylab("-log10(Pvalue)") +               # 设置y轴标签  scale_color_manual(values = c("blue4", "grey", "red3"))p2    

#增加区域划分并改变主题p3 <- ggplot(data = dt,            aes(x = logFC,                y = -log10(Pval))) +  # 设置x轴为logFC，y轴为-P.Value的对数值  geom_point(alpha = 0.4, size = 2.5,             aes(color = catogrey)) +      # 添加散点，根据change列着色  ylab("-log10(Pvalue)") +               # 设置y轴标签  scale_color_manual(values = c("blue4", "grey", "red3")) +  # 设置颜色映射，蓝色表示下调，灰色表示稳定，红色表示上调  geom_vline(xintercept = c(-2, 2), lty = 4, col = "black", lwd = 0.8) +  # 添加垂直参考线，用于标记logFC阈值  geom_hline(yintercept = -log10(0.001), lty = 4, col = "black", lwd = 0.8) +   # 添加水平参考线，用于标记-P.Value阈值      theme_bw() + theme(panel.grid=element_blank())  # 使用网格白底主题p3

#########对符合筛选条件的基因添加标签#筛选上调中差异倍数最大的10个基因up_list <- dt %>%  filter(catogrey == 'up') %>%  distinct(name,.keep_all = T) %>%  #其实如果像是我这种没有具体的gene symbol的数据是不用这一步的  top_n(10,logFC)    head(up_list)

#筛选下调中差异倍数最大的10个基因down_list <- dt %>%  filter(catogrey == 'down') %>%  distinct(name,.keep_all = T) %>%  top_n(10,-logFC)head(down_list)

#对火山图添加标签P4 <- p3 + geom_text_repel(data = up_list, aes(x = logFC, y = -log10(Pval), label = name)) +  # 添加上调基因的标签  geom_text_repel(data = down_list, aes(x = logFC, y = -log10(Pval), label = name))  # 添加下调基因的标签P4    

#改变标签样式p5 <- p3 +  geom_label_repel(data = up_list, aes(x = logFC, y = -log10(Pval), label = name)) +  # 添加上调基因的标签  geom_label_repel(data = down_list, aes(x = logFC, y = -log10(Pval), label = name))p5   

#大家可以看到，在进行以上标签时，的确会报warnning，但是#在对需要注释的点重新绘制点就不会出现这个错误p6 <- p3 +  # 基于普通火山图p  geom_point(data = up_list,                             # 上调数据集             aes(x = logFC, y = -log10(Pval)),             color = 'red3', size = 4.5, alpha = 0.2) +  # 三点颜色应与p3相同  geom_label_repel(data = up_list,                       # 添加标签                   aes(x = logFC, y = -log10(Pval), label = name),                       seed = 233,                           # 随机数种子，有过介绍哦                   size = 2.5,                           # 字体大小改变                   color = 'black',                      # 标签的字体颜色                   min.segment.length = 0,               # 是否始终添加引线                   force = 2,                            # 若标签重叠，则出现排斥                   force_pull = 2,                       # 标签与数据点间的吸引力，可以自己改变试一下                   box.padding = 0.4,                    # 标签周围空间                   max.overlaps = Inf) +                  # 保持始终显示所有标签  geom_point(data = down_list,                                         aes(x = logFC, y = -log10(Pval)),             color = 'blue4', size = 4.5, alpha = 0.2) +   geom_label_repel(data = down_list,                                        aes(x = logFC, y = -log10(Pval), label = name),                   seed = 233,                                             size = 2.5,                                           color = 'black',                                       min.segment.length = 0,                                 force = 2,                                              force_pull = 2,                                         box.padding = 0.4,                                      max.overlaps = Inf)                       p6