免疫治疗结果与风险关联的直方图

1.需求

为了直观的展示免疫治疗结果与病人风险高低之间的关系，如图：

2.数据

需要病人的高低风险分组，和免疫治疗结果

set.seed(10086)
meta = data.frame(Sample = paste0("sample",1:100),
                  Response = sample(c("SD/PD","CR/PR"),100,replace = T),
                  risk = sample(c("High","Low"),100,replace = T))
head(meta)

##    Sample Response risk
## 1 sample1    CR/PR  Low
## 2 sample2    CR/PR High
## 3 sample3    SD/PD  Low
## 4 sample4    SD/PD  Low
## 5 sample5    SD/PD  Low
## 6 sample6    CR/PR High

3.画图代码

library(dplyr)
dat = count(meta,risk,Response)
dat = dat %>% group_by(risk) %>% 
  summarise(Response = Response,n = n/sum(n))
dat$Response = factor(dat$Response,levels = c("SD/PD","CR/PR"))
dat

## # A tibble: 4 x 3
## # Groups:   risk [2]
##   risk  Response     n
##   <chr> <fct>    <dbl>
## 1 High  CR/PR    0.628
## 2 High  SD/PD    0.372
## 3 Low   CR/PR    0.474
## 4 Low   SD/PD    0.526

library(ggplot2)
ggplot(data = dat)+
  geom_bar(aes(x = risk,y = n,
               fill = Response),
           stat = "identity")+
  scale_fill_manual(values =  c("#f87669","#2fa1dd"))+
  geom_text(aes(x = risk,y = n,
                label = scales::percent(n)),
            color = "white",
            size = 6,
            position = position_fill(vjust = 0.5))+
  theme_minimal()+
  theme(legend.position = "top")

因为数据是编的，就不用看high和low两组免疫治疗结果好坏所占的比例啦。实战数据再看即可。