跟着Science学画图：R语言ggplot2作热图展示基因存在缺失变异（PAV）

df<-read.csv("numeric_pan_matrix.csv",
             row.names = 1)
dim(df)

library(dplyr)


df %>% 
  mutate(x=1:nrow(.)) %>% 
  select(2:27,x) %>% 
  reshape2::melt(,id.vars="x") %>% 
  mutate(pav=case_when(
    value == 0 ~ "Absence",
    TRUE ~ "Presence"
  )) -> dfa

dfa %>% count(pav)

pdf(file = "pavheatmap.pdf",
    width = 10,
    height = 6,
    family = "serif")
ggplot(data=dfa,aes(x=x,y=variable))+
  geom_tile(aes(fill=pav))+
  scale_fill_manual(name="Pan-Genes (n=103033) Presence or Absence",
                    values=c("#316879","#ff9a8d"))  + 
  theme_classic() + 
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(), 
        axis.ticks.x=element_blank()) +
  theme(axis.title.y=element_blank(),
        axis.text.y=element_text(angle = 0, 
                                 hjust = 1, 
                                 colour = color_fill),
        axis.ticks.y=element_blank())  +
  theme(legend.position="bottom") +
  theme(text = element_text(size = 12)) 

dev.off()