如何用多因子表绘制多条曲线？

Question

我有一张像这样的桌子

Condition   downsampling    E. coli S. cerevisiae
 Treated    45000000    1   0.944968385
 Treated    40000000    1   0.932060195
 Treated    32000000    1   0.900323585
 Treated    16000000    0.99999549  0.73127366
 Treated    8000000 0.99993898  0.503170515
 Treated    4000000 0.99892133  0.30287704
 Treated    2000000 0.97810106  0.16861184
 Treated    1000000 0.86656028  0.089200035
Untreated   45000000    0.886457145 0.108071345
Untreated   40000000    0.85728706  0.09729946
Untreated   32000000    0.79344402  0.08072991
Untreated   16000000    0.553520285 0.04306675
Untreated   8000000 0.337149605 0.023035225
Untreated   4000000 0.18756713  0.0119472
Untreated   2000000 0.097686445 0.006072755
Untreated   1000000 0.05007619  0.0031243

如何在 R 中生成一个图表，其中有 4 条曲线，分别代表每种细菌的两个条件？请参阅附上 Excel 中生成的图片以了解我想要获得的内容

沿着这些路线尝试了 ggplot 的一些东西，但不起作用

test <- ggplot(dfuse, aes(x=Downsampling))+
  geom_line(data = dfuse[dfuse$Treatment=="Untreated",], aes(x=Downsampling, y=E.coli, color="E.coli untreated"), size=1.5) +
  geom_line(data = dfuse[dfuse$Treatment=="Untreated",], aes(x=Downsampling, y=S.cerevisiae, color="S.cerevisiae untreated"), size=1.5) +
  geom_line(data = dfuse[dfuse$Treatment=="ALU Plus 120K 1hr 1X 42C",], aes(x=Downsampling, y=E.coli, color="E.coli treated"), size=1.5) +
  geom_line(data = dfuse[dfuse$Treatment=="ALU Plus 120K 1hr 1X 42C",], aes(x=Downsampling, y=S.cerevisiae, color="S.cerevisiae treated"), size=1.5) +
  labs(x="Downsampling", y="Genome Coverage (%)") +
  scale_color_manual(values=c("E.coli untreated"="blue", "S.cerevisiae untreated"="red", "E.coli treated"="green", "S.cerevisiae treated"="purple")) +
  theme_minimal()
  
test
  
ggsave(test,filename = "test.png", height=8, width=10)

Answer 1

library(dplyr)
library(tidyr)
library(ggplot2)

dfuse %>% 
  pivot_longer(-c(Condition, downsampling)) %>% 
  mutate(cb = paste(gsub("\\.\\.", "\\.", name), tolower(Condition))) %>% 
  ggplot(aes(x = downsampling/1e6, y = value, color = cb, group = cb)) + 
  geom_line(color = "black") +
  geom_point() + 
  labs(title = "Microbial genome breradth of coverage",
       x="Downsampling (Millions)", y="Genome Coverage (%)") +
  scale_x_continuous(limits = c(0, 50)) +
  scale_y_continuous(labels = scales::percent) +
  scale_color_manual(values=c("E.coli untreated"="blue", 
                              "S.cerevisiae untreated"="red", 
                              "E.coli treated"="green", 
                              "S.cerevisiae treated"="purple"),
                    name = "") +
  theme_classic() +
  theme(legend.position="bottom",
        plot.title = element_text(hjust = 0.5))

数据：

read.table(text = "Condition   downsampling    'E. coli' 'S. cerevisiae'
 Treated    45000000    1   0.944968385
 Treated    40000000    1   0.932060195
 Treated    32000000    1   0.900323585
 Treated    16000000    0.99999549  0.73127366
 Treated    8000000 0.99993898  0.503170515
 Treated    4000000 0.99892133  0.30287704
 Treated    2000000 0.97810106  0.16861184
 Treated    1000000 0.86656028  0.089200035
Untreated   45000000    0.886457145 0.108071345
Untreated   40000000    0.85728706  0.09729946
Untreated   32000000    0.79344402  0.08072991
Untreated   16000000    0.553520285 0.04306675
Untreated   8000000 0.337149605 0.023035225
Untreated   4000000 0.18756713  0.0119472
Untreated   2000000 0.097686445 0.006072755
Untreated   1000000 0.05007619  0.0031243", 
header = T, stringsAsFactors = T) -> dfuse

^{创建于 2024-03-28，使用 reprex v2.0.2}