根据x轴名称所属组的字母顺序对其重新排序

Question

我有一个数据框：

gene_symbol<-c("DADA","SDAASD","SADDSD","SDADD","ASDAD","XCVXCVX","EQWESDA","DASDADS","SDASDASD","DADADASD","sdaadfd","DFSD","SADADDAD","SADDADADA","DADSADSASDWQ","SDADASDAD","ASD","DSADD")
panel<-c("growth","growth","growth","growth","big","big","big","small","small","dfgh","DF","DF","DF","DF","DF","gh","DF","DF")
ASDDA<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDb<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf1<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf2<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf3<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf4<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf5<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDA1<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDb1<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf1<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf11<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf21<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf31<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf41<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
ASDDAf51<-c("normal","over","low","over","normal","over","low","over","normal","over","DF","DF","DF","DF","DF","DF","DF","DF")
Gene_states22<-data.frame(gene_symbol,panel,ASDDA,ASDDb,ASDDAf,ASDDAf1,ASDDAf2,ASDDAf3,ASDDAf4,ASDDAf5,ASDDA1,ASDDb1,ASDDAf1,ASDDAf11,ASDDAf21,ASDDAf31,ASDDAf41,ASDDAf51)

and I create a heatmap:

library(ggplot2); library(reshape2)
HG3 <- split(Gene_states22, Gene_states22$panel)
HG4 <- melt(HG3, id.vars= c("gene_symbol","panel"))
HG4 <- HG4[,-5]

pp <- ggplot(HG4, aes(gene_symbol,variable)) + 
  geom_tile(aes(fill = value),
            colour = "grey50") + 
  scale_fill_manual(values = c("white", "red", "blue", "black", "yellow", "green", "brown")) +
  labs(title = "Heatmap", x = "gene_symbol", y = "sample", fill = "value") +
  guides(fill = FALSE)+
  theme(panel.background = element_rect(fill = NA),
        panel.spacing = unit(0.5, "lines"), ## It was here where you had a 0 for distance between facets. I replaced it by 0.5 .
        strip.placement = "outside")
pp

您将看到x轴上的名称按字母顺序排列。我想根据它们所属的组（panel）的字母顺序对其重新排序。因此，首先将显示属于“ big”的名称，依此类推。

Answer 1

代码的两个小补充（带有注释）：

library(ggplot2)
library(reshape2)

HG3 <- split(Gene_states22, Gene_states22$panel)
HG4 <- melt(HG3, id.vars= c("gene_symbol","panel"))
HG4 <- HG4[,-5]

# reset your factor levels based on how they appear in your dataset and not alphabetically
HG4$gene_symbol = factor(HG4$gene_symbol, levels = unique(HG4$gene_symbol))

pp <- ggplot(HG4, aes(gene_symbol,variable)) + 
  geom_tile(aes(fill = value),
            colour = "grey50") + 
  scale_fill_manual(values = c("white", "red", "blue", "black", "yellow", "green", "brown")) +
  labs(title = "Heatmap", x = "gene_symbol", y = "sample", fill = "value") +
  guides(fill = FALSE)+
  theme(panel.background = element_rect(fill = NA),
        panel.spacing = unit(0.5, "lines"), 
        strip.placement = "outside",
        axis.text.x = element_text(angle = 45, hjust = 1))  # add this to have a better visualisation (uses an angle to avoid overlap of text)
pp

Answer 2

您可以创建一个数字分组变量，该变量按字母顺序排序（使用URL Python）；然后使用dplyr's包中的group_indices对x轴进行相应排序。

fct_reorder

-

forcats