GOplot｜宝藏R包，拯救你的GO富集结果，杜绝平庸的条形图

前言

今天给大家分享一个宝藏R包：GOplot

包如其名，专作GO富集分析的可视化工具包。函数网上教程不少，但这里笔者已经列出了所有可修改的常用参数，改色改大小排序自定义，整整齐齐。

百行不到的代码，直接上升一个level：

数据准备

下载包，加载数据，这里要留意数据的格式，这是把差异分析的结果也纳入了的

# install.packages('GOplot')
library("GOplot")
data(EC)
head(EC$david)
head(EC$genelist)
circ <- circle_dat(EC$david, EC$genelist)

GOBar

若把order.by.zscore修改为T则改为取绝对值进行排序

GOBar(circ, 
      # category == 'BP',
      display = 'multiple', # single
      order.by.zscore = F, #order
      title = 'Z-score coloured barplot',
      zsc.col = c('yellow', 'black', 'cyan'))

GOBubble

进行了一步过滤，以减少重叠。若不过滤可直接将数据改为circ

# 筛选重叠
reduced_circ <- reduce_overlap(circ, overlap = 0.75)
GOBubble(reduced_circ, # circ
         title = 'Bubble plot', 
         colour = c("#4DBBD5FF","#EFD500FF","#F39B7FFF"), 
         table.legend = T,
         bg.col = T,
         ID = T, # GO id
         display = 'multiple', 
         labels = 3)

GOCircle

IDs <- c('GO:0007507', 'GO:0001568', 'GO:0001944', 'GO:0048729', 'GO:0048514', 'GO:0005886', 'GO:0008092', 'GO:0008047')
GOCircle(circ, nsub = IDs,
         rad1 = 2, 
         rad2 = 3,
         table.legend = T,
         zsc.col = c('darkgoldenrod1', 'black', 'cyan1'),
         lfc.col = c("#F39B7FFF","#4DBBD5FF"),
         label.size = 5,
         label.fontface = "bold")

GOChord

这里需要把数据转换成长格式

chord <- chord_dat(data = circ, genes = EC$genes, process = EC$process)
GOChord(chord,
        title = "GOChord Plot",
        space = 0.02, 
        gene.order = 'logFC', 
        gene.space = 0.25, 
        gene.size = 3,
        nlfc = 1,
        lfc.col = c('darkgoldenrod1', 'black', 'cyan1'),
        lfc.min = -4,
        lfc.max = 4,
        border.size = 0.5,
        process.label = 8)

GOHeat

GOHeat(chord[,-8], # 剔除FC列
       nlfc = 0)

GOHeat(chord, 
       nlfc = 1, # 倒数第一列
       fill.col = c("#4DBBD5FF","#EFD500FF","#F39B7FFF"))

第一个是剔除FC列，默认统计富集的功能数量

第二个是保留FC列，指定参照倒数第一列差异倍数

GOCluster

这里是按FC聚类，可以将clust.by指定成按term聚类

GOCluster(circ, 
          EC$process, 
          metric = "euclidean", # 欧式距离
          clust.by = 'logFC', # 'term'
          nlfc = F, # T为多列FC，F则为一列FC
          lfc.col = c('darkgoldenrod1', 'black', 'cyan1'),
          lfc.min = -3,
          lfc.max = 3,
          lfc.space = 0,
          lfc.width = 0.5,
          term.col = ggsci::pal_frontiers(alpha = 0.7)(7), # max = 20
          term.space = 0.5,
          term.width = 2)

GOVenn

红色为上调基因个数，紫色为下调基因个数，灰色则为交集后上下调冲突的基因个数

l1 <- subset(circ, term == 'heart development', c(genes,logFC))
l2 <- subset(circ, term == 'plasma membrane', c(genes,logFC))
l3 <- subset(circ, term == 'tissue morphogenesis', c(genes,logFC))
GOVenn(l1,l2,l3, 
       label = c('heart development', 'plasma membrane', 'tissue morphogenesis'),
       lfc.col = c("#EE0000FF","#808180FF","#5F559BFF"),
       circle.col = c("#EFD500FF","#4DBBD5FF","#F39B7FFF"),
       plot = T)

shiny

这里作者还做了个venn的shiny交互网站，有兴趣的小伙伴可以试一试：https://wwalter.shinyapps.io/Venn/

小总结

告别ggplot2手画，GO富集的红蓝气泡图、柱状图已经审美疲劳了。这个GOplot应该是懒癌晚期拯救者了，都是打包好的函数直接换数据运行即可，记得文献引用上：Walter, Wencke, Fátima Sánchez-Cabo, and Mercedes Ricote. "GOplot: an R package for visually combining expression data with functional analysis." Bioinformatics (2015): btv300.