在`R`中的`gplots`中，如何调整`heatmap.2`函数中x-label ticks的位置？

Question

我正在尝试复制在pg上发现的here的情节。4：

​

​

它的可重现代码是：

require(devtools)
install_git("https://github.com/marchion/git.switchBox", subdir="switchBox")
require(switchBox)
require(gplots)
data(trainingData)    
classifier <- SWAP.KTSP.Train(matTraining, trainingGroup)
kappa <- SWAP.KTSP.Statistics(matTraining, classifier)
mat <- t(1*kappa$comparisons)
rownames(mat) <- gsub(">", "\n more express than\n", rownames(mat))

heatmap.2(mat,
scale="none", Rowv=F, Colv=F, dendrogram="none",
trace="none", key=FALSE,
col=c("lightsteelblue2", "pink3"),
labCol=toupper(paste(trainingGroup, "Prognosis")),
sepwidth=c(0.075,0.075), sepcolor="black",
rowsep=1:ncol(kappa$comparisons),
colsep=1:nrow(kappa$comparisons),
lmat=rbind( c(0, 3), c(2, 1), c(0, 4) ), lhei=c(0.1, 5, 0.5), lwid=c(0.15, 5),
mar=c(7.5, 12), cexRow=0.85, cexCol=0.9)

如果您在上面的图中注意到，x标签稍微向左偏离中心。在heatmap.2函数中是否有一个命令可以将每个标签向右移动？

Answer 1

你必须指定参数adjCol (c(1, 0.5))会给你想要的结果(c(1, 0)会把它移到左边，c(1, 1)会把它移到右边)。

代码(使用OPs提供的包和数据)：

heatmap.2(
  mat,
  adjCol = c(1, 0.5), 
  scale = "none", Rowv = FALSE, Colv = FALSE, dendrogram = "none",
  trace = "none", key = FALSE,
  col = c("lightsteelblue2", "pink3"),
  labCol = toupper(paste(trainingGroup, "Prognosis")),
  sepwidth = c(0.075,0.075), sepcolor = "black",
  rowsep = 1:ncol(kappa$comparisons),
  colsep = 1:nrow(kappa$comparisons),
  lmat = rbind( c(0, 3), c(2, 1), c(0, 4) ),
  lhei = c(0.1, 5, 0.5), lwid = c(0.15, 5),
  mar = c(7.5, 12), cexRow = 0.85, cexCol = 0.9,
)

结果：

​

​