如何在 ggplot2 的图例中插入表达式？:: 正确的颜色 + 多条线和点
How Insert an expression in legend in ggplot2?:: correct color + multiple lines and point
我是 R 的新手，无法更正下图。
Xb_exp，它应该有蓝点。
Xb_dw，红色实线。
Xb_f，虚线。
Xb_s，长划线。
图例表达式应如带下标所示。
我无法纠正它。
有办法吗？
enter image description here
my data
CA <- c(3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30)
Xb_exp <- c(0.0231,0.0519,0.0839,0.1197,0.1595,0.1996,0.2384,0.2772,0.3153,0.3520,0.3887,0.4254,0.4615,0.4978,0.5339,0.5685,0.6000,0.6279,0.6528,0.6762,0.6974,0.7166,0.7346,0.7516,0.7669,0.7810,0.7940,0.8059)
Xb_dw <- c(0.0160,0.0516,0.0886,0.1259,0.1633,0.2006,0.2377,0.2749,0.3122,0.3496,0.3870,0.4245,0.4617,0.4984,0.5339,0.5678,0.5996,0.6288,0.6551,0.6786,0.6994,0.7179,0.7346,0.7499,0.7641,0.7774,0.7899,0.8018)
Xb_f <- c(0.0021,0.0031,0.0046,0.0067,0.0095,0.0131,0.0177,0.0234,0,0387,0.0483,0.0591,0.0709,0.0832,0.0955,0.1073,0.1181,0.1272,0.1345,0.1398,0.1443,0.1456,0.1468,0.1474,0.1476,0.1477,0.1477,0.1477,0.1477)
Xb_s <- c(0.0139,0.0484,0.0839,0.1192,0.1538,0.1874,0.2200,0.2515,0.2818,0.3108,0.3387,0.3653,0.3908,0.4151,0.4383,0.4604,0.4815,0.5015,0.5206,0.5387,0.5559,0.5722,0.5877,0.6024,0.6164,0.6264,0.6421,0.6040)
dat <- c(CA, Xb_exp, Xb_dw, Xb_f, Xb_s)
my code
labels = c(expression(X[b_exp]),expression(X[b_dw]),expression(X[b_f]),expression(X[b_s]))
color4 <- c("Xb_exp"="#3C5488FF", "Xb_dw"="#DC0000FF", "Xb_f"="#00A087FF", "Xb_s"="#4DBBD5FF")
Xb_D1 <- ggplot(data = dat) +
theme_bw() +
labs(x="Crank position (ºCA)", y= bquote('Burn fraction ('~X[b]~')')) +
geom_point(aes(x=CA, y=Xb_exp, colour="Xb_exp"), size=3) +
geom_line(aes(x=CA, y=Xb_dw,colour="Xb_dw"), size=1,linetype="solid") +
geom_line(aes(x=CA, y=Xb_f,colour="Xb_f"), size=1,linetype="dotted") +
geom_line(aes(x=CA, y=Xb_s,colour="Xb_s"), size=1,linetype="longdash") +
scale_colour_manual(values=color4, labels=labels) +
theme(legend.title = element_blank(),legend.position = c(0.8, 0.5),
legend.text = element_text(size = 12)) +
scale_x_continuous(limits = c(2,80))
plot(Xb_D1)
ggplot() 需要数据框而不是向量。如果您修改代码：
dat <- data.frame(CA, Xb_exp, Xb_dw, Xb_f, Xb_s)
并更正 Xb_f 矢量中的拼写错误
Xb_f <- c(0.0021,0.0031,0.0046,0.0067,0.0095,0.0131,0.0177,0.0234,0.0387,0.0483,0.0591,0.0709,0.0832,0.0955,0.1073,0.1181,0.1272,0.1345,0.1398,0.1443,0.1456,0.1468,0.1474,0.1476,0.1477,0.1477,0.1477,0.1477)
您的剩余代码将正常工作，但可以使用下面的 tidyverse 方法更简单地实现。使用 pivot_longer 将 y 变量与 x 变量叠加。
dat %>%
pivot_longer(Xb_exp:Xb_s) %>%
ggplot(aes(x = CA, y = value, colour = name)) +
geom_point() +
geom_line() +
scale_colour_manual(values=color4, labels=labels) +
theme_bw() +
theme(legend.title = element_blank(),legend.position = c(0.8, 0.5),
legend.text = element_text(size = 12)) +
scale_x_continuous(limits = c(2,80)) +
labs(x="Crank position (ºCA)", y= bquote('Burn fraction ('~X[b]~')')) ```
您的代码中存在一些错误，表明您没有尝试粘贴的内容。
0,0387, 在你的数据中应该是 0.0387, 否则没有什么是正确的（那里没有数据测量几百）c(CA, ... ) 应该是 data.frame( CA, ... )
现在，第一个问题是您自己完成了所有繁重的工作，而 ggplot 坐在那里无所事事。它旨在按组设置颜色和线型。但是，您需要先转换数据才能充分利用它：
library(tidyr)
CA <- c(3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30)
Xb_exp <- c(0.0231,0.0519,0.0839,0.1197,0.1595,0.1996,0.2384,0.2772,0.3153,0.3520,0.3887,0.4254,0.4615,0.4978,0.5339,0.5685,0.6000,0.6279,0.6528,0.6762,0.6974,0.7166,0.7346,0.7516,0.7669,0.7810,0.7940,0.8059)
Xb_dw <- c(0.0160,0.0516,0.0886,0.1259,0.1633,0.2006,0.2377,0.2749,0.3122,0.3496,0.3870,0.4245,0.4617,0.4984,0.5339,0.5678,0.5996,0.6288,0.6551,0.6786,0.6994,0.7179,0.7346,0.7499,0.7641,0.7774,0.7899,0.8018)
Xb_f <- c(0.0021,0.0031,0.0046,0.0067,0.0095,0.0131,0.0177,0.0234,0.0387,0.0483,0.0591,0.0709,0.0832,0.0955,0.1073,0.1181,0.1272,0.1345,0.1398,0.1443,0.1456,0.1468,0.1474,0.1476,0.1477,0.1477,0.1477,0.1477)
Xb_s <- c(0.0139,0.0484,0.0839,0.1192,0.1538,0.1874,0.2200,0.2515,0.2818,0.3108,0.3387,0.3653,0.3908,0.4151,0.4383,0.4604,0.4815,0.5015,0.5206,0.5387,0.5559,0.5722,0.5877,0.6024,0.6164,0.6264,0.6421,0.6040)
dat <- data.frame(CA, Xb_exp, Xb_dw, Xb_f, Xb_s)
color4 <- c("Xb_exp"="#3C5488FF", "Xb_dw"="#DC0000FF", "Xb_f"="#00A087FF", "Xb_s"="#4DBBD5FF")
linetypes <- c( Xb_dw="solid", Xb_f="dotted", Xb_s="longdash" )
dat2 <- pivot_longer( dat, cols=starts_with("Xb_") )
dat2.line <- dat2 %>% filter( name != "Xb_exp" )
dat2.point <- dat2 %>% filter( name == "Xb_exp" )
dat2 现在是一个 long 数据集，数据类别作为变量，而不是每个数据系列单独列。这就是 ggplot 喜欢它的方式：
dat2
# A tibble: 112 x 3
CA name value
<dbl> <fct> <dbl>
1 3 Xb_exp 0.0231
2 3 Xb_dw 0.016
3 3 Xb_f 0.0021
4 3 Xb_s 0.0139
5 4 Xb_exp 0.0519
6 4 Xb_dw 0.0516
7 4 Xb_f 0.0031
8 4 Xb_s 0.0484
9 5 Xb_exp 0.0839
10 5 Xb_dw 0.0886
# … with 102 more rows
然后我将数据拆分成点和线，只是为了不让情节代码比它必须的更难看：
Xb_D1 <- ggplot(data = dat2.line, aes(x=CA,y=value,color=name)) +
theme_bw() +
labs(x="Crank position (ºCA)", y= bquote('Burn fraction ('~X[b]~')')) +
geom_point( data = dat2.point, size=3) +
geom_line( aes(col=name,lty=name), size=1 ) +
scale_colour_manual(values=color4) +
scale_linetype_manual( values=linetypes, guide=FALSE ) +
guides(
color = guide_legend( override.aes=list( shape=c(NA,19,NA,NA), linetype=c("solid","solid","dashed","dotted") ) )
) +
theme(legend.title = element_blank(),legend.position = c(0.8, 0.5),
legend.text.align = 0,
legend.text = element_text(size = 12)) +
scale_x_continuous(limits = c(2,30))
print(Xb_D1)
- 无需提供标签
- 使用线型就像在 ggplot 中使用颜色一样，它只是另一个可以承载信息的通道（或者 aesthetic，因为他们喜欢在那里称呼它）
- 将图例左对齐，那样看起来更好
- 更复杂的是使用override.aes从图例类别中拿走不应该拥有它们的点数。
现在，我无法更改标签中数据系列的顺序，这可能会很麻烦。他们现在的顺序对你来说还可以吗？
具有讽刺意味的是，使用常规绘图进行设置非常简单：
鉴于以上所有数据：
linetypes4 <- c( Xb_exp=NA, Xb_dw="solid", Xb_f="dotted", Xb_s="longdash" )
plot(
NA, type="n", xlim=c(0,30), ylim=c(0,0.8),
xlab = "Crank position (ºCA)", ylab = bquote('Burn fraction ('~X[b]~')'),
panel.first = grid()
)
with( dat, {
points( x=CA, y=Xb_exp, pch=19, col=color4["Xb_exp"], size=3 )
for( n in c("Xb_dw", "Xb_f", "Xb_s")) {
lines( x=CA, y=get(n), lty=linetypes[n], col=color4[n], lwd=2 )
}
})
legend(
x = "right",
legend = labels,
col = color4,
lty = linetypes4,
pch = c(19,NA,NA,NA),
box.lwd = 0,
inset = .02
)
我是 R 的新手，无法更正下图。 Xb_exp，它应该有蓝点。 Xb_dw，红色实线。 Xb_f，虚线。 Xb_s，长划线。 图例表达式应如带下标所示。 我无法纠正它。 有办法吗？
enter image description here
my data
CA <- c(3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30)
Xb_exp <- c(0.0231,0.0519,0.0839,0.1197,0.1595,0.1996,0.2384,0.2772,0.3153,0.3520,0.3887,0.4254,0.4615,0.4978,0.5339,0.5685,0.6000,0.6279,0.6528,0.6762,0.6974,0.7166,0.7346,0.7516,0.7669,0.7810,0.7940,0.8059)
Xb_dw <- c(0.0160,0.0516,0.0886,0.1259,0.1633,0.2006,0.2377,0.2749,0.3122,0.3496,0.3870,0.4245,0.4617,0.4984,0.5339,0.5678,0.5996,0.6288,0.6551,0.6786,0.6994,0.7179,0.7346,0.7499,0.7641,0.7774,0.7899,0.8018)
Xb_f <- c(0.0021,0.0031,0.0046,0.0067,0.0095,0.0131,0.0177,0.0234,0,0387,0.0483,0.0591,0.0709,0.0832,0.0955,0.1073,0.1181,0.1272,0.1345,0.1398,0.1443,0.1456,0.1468,0.1474,0.1476,0.1477,0.1477,0.1477,0.1477)
Xb_s <- c(0.0139,0.0484,0.0839,0.1192,0.1538,0.1874,0.2200,0.2515,0.2818,0.3108,0.3387,0.3653,0.3908,0.4151,0.4383,0.4604,0.4815,0.5015,0.5206,0.5387,0.5559,0.5722,0.5877,0.6024,0.6164,0.6264,0.6421,0.6040)
dat <- c(CA, Xb_exp, Xb_dw, Xb_f, Xb_s)
my code
labels = c(expression(X[b_exp]),expression(X[b_dw]),expression(X[b_f]),expression(X[b_s]))
color4 <- c("Xb_exp"="#3C5488FF", "Xb_dw"="#DC0000FF", "Xb_f"="#00A087FF", "Xb_s"="#4DBBD5FF")
Xb_D1 <- ggplot(data = dat) +
theme_bw() +
labs(x="Crank position (ºCA)", y= bquote('Burn fraction ('~X[b]~')')) +
geom_point(aes(x=CA, y=Xb_exp, colour="Xb_exp"), size=3) +
geom_line(aes(x=CA, y=Xb_dw,colour="Xb_dw"), size=1,linetype="solid") +
geom_line(aes(x=CA, y=Xb_f,colour="Xb_f"), size=1,linetype="dotted") +
geom_line(aes(x=CA, y=Xb_s,colour="Xb_s"), size=1,linetype="longdash") +
scale_colour_manual(values=color4, labels=labels) +
theme(legend.title = element_blank(),legend.position = c(0.8, 0.5),
legend.text = element_text(size = 12)) +
scale_x_continuous(limits = c(2,80))
plot(Xb_D1)
ggplot() 需要数据框而不是向量。如果您修改代码：
dat <- data.frame(CA, Xb_exp, Xb_dw, Xb_f, Xb_s)
并更正 Xb_f 矢量中的拼写错误
Xb_f <- c(0.0021,0.0031,0.0046,0.0067,0.0095,0.0131,0.0177,0.0234,0.0387,0.0483,0.0591,0.0709,0.0832,0.0955,0.1073,0.1181,0.1272,0.1345,0.1398,0.1443,0.1456,0.1468,0.1474,0.1476,0.1477,0.1477,0.1477,0.1477)
您的剩余代码将正常工作，但可以使用下面的 tidyverse 方法更简单地实现。使用 pivot_longer 将 y 变量与 x 变量叠加。
dat %>%
pivot_longer(Xb_exp:Xb_s) %>%
ggplot(aes(x = CA, y = value, colour = name)) +
geom_point() +
geom_line() +
scale_colour_manual(values=color4, labels=labels) +
theme_bw() +
theme(legend.title = element_blank(),legend.position = c(0.8, 0.5),
legend.text = element_text(size = 12)) +
scale_x_continuous(limits = c(2,80)) +
labs(x="Crank position (ºCA)", y= bquote('Burn fraction ('~X[b]~')')) ```
您的代码中存在一些错误，表明您没有尝试粘贴的内容。
0,0387,在你的数据中应该是0.0387,否则没有什么是正确的（那里没有数据测量几百）c(CA, ... )应该是data.frame( CA, ... )
现在，第一个问题是您自己完成了所有繁重的工作，而 ggplot 坐在那里无所事事。它旨在按组设置颜色和线型。但是，您需要先转换数据才能充分利用它：
library(tidyr)
CA <- c(3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30)
Xb_exp <- c(0.0231,0.0519,0.0839,0.1197,0.1595,0.1996,0.2384,0.2772,0.3153,0.3520,0.3887,0.4254,0.4615,0.4978,0.5339,0.5685,0.6000,0.6279,0.6528,0.6762,0.6974,0.7166,0.7346,0.7516,0.7669,0.7810,0.7940,0.8059)
Xb_dw <- c(0.0160,0.0516,0.0886,0.1259,0.1633,0.2006,0.2377,0.2749,0.3122,0.3496,0.3870,0.4245,0.4617,0.4984,0.5339,0.5678,0.5996,0.6288,0.6551,0.6786,0.6994,0.7179,0.7346,0.7499,0.7641,0.7774,0.7899,0.8018)
Xb_f <- c(0.0021,0.0031,0.0046,0.0067,0.0095,0.0131,0.0177,0.0234,0.0387,0.0483,0.0591,0.0709,0.0832,0.0955,0.1073,0.1181,0.1272,0.1345,0.1398,0.1443,0.1456,0.1468,0.1474,0.1476,0.1477,0.1477,0.1477,0.1477)
Xb_s <- c(0.0139,0.0484,0.0839,0.1192,0.1538,0.1874,0.2200,0.2515,0.2818,0.3108,0.3387,0.3653,0.3908,0.4151,0.4383,0.4604,0.4815,0.5015,0.5206,0.5387,0.5559,0.5722,0.5877,0.6024,0.6164,0.6264,0.6421,0.6040)
dat <- data.frame(CA, Xb_exp, Xb_dw, Xb_f, Xb_s)
color4 <- c("Xb_exp"="#3C5488FF", "Xb_dw"="#DC0000FF", "Xb_f"="#00A087FF", "Xb_s"="#4DBBD5FF")
linetypes <- c( Xb_dw="solid", Xb_f="dotted", Xb_s="longdash" )
dat2 <- pivot_longer( dat, cols=starts_with("Xb_") )
dat2.line <- dat2 %>% filter( name != "Xb_exp" )
dat2.point <- dat2 %>% filter( name == "Xb_exp" )
dat2 现在是一个 long 数据集，数据类别作为变量，而不是每个数据系列单独列。这就是 ggplot 喜欢它的方式：
dat2
# A tibble: 112 x 3
CA name value
<dbl> <fct> <dbl>
1 3 Xb_exp 0.0231
2 3 Xb_dw 0.016
3 3 Xb_f 0.0021
4 3 Xb_s 0.0139
5 4 Xb_exp 0.0519
6 4 Xb_dw 0.0516
7 4 Xb_f 0.0031
8 4 Xb_s 0.0484
9 5 Xb_exp 0.0839
10 5 Xb_dw 0.0886
# … with 102 more rows
然后我将数据拆分成点和线，只是为了不让情节代码比它必须的更难看：
Xb_D1 <- ggplot(data = dat2.line, aes(x=CA,y=value,color=name)) +
theme_bw() +
labs(x="Crank position (ºCA)", y= bquote('Burn fraction ('~X[b]~')')) +
geom_point( data = dat2.point, size=3) +
geom_line( aes(col=name,lty=name), size=1 ) +
scale_colour_manual(values=color4) +
scale_linetype_manual( values=linetypes, guide=FALSE ) +
guides(
color = guide_legend( override.aes=list( shape=c(NA,19,NA,NA), linetype=c("solid","solid","dashed","dotted") ) )
) +
theme(legend.title = element_blank(),legend.position = c(0.8, 0.5),
legend.text.align = 0,
legend.text = element_text(size = 12)) +
scale_x_continuous(limits = c(2,30))
print(Xb_D1)
- 无需提供标签
- 使用线型就像在 ggplot 中使用颜色一样，它只是另一个可以承载信息的通道（或者 aesthetic，因为他们喜欢在那里称呼它）
- 将图例左对齐，那样看起来更好
- 更复杂的是使用override.aes从图例类别中拿走不应该拥有它们的点数。
现在，我无法更改标签中数据系列的顺序，这可能会很麻烦。他们现在的顺序对你来说还可以吗？
具有讽刺意味的是，使用常规绘图进行设置非常简单：
鉴于以上所有数据：
linetypes4 <- c( Xb_exp=NA, Xb_dw="solid", Xb_f="dotted", Xb_s="longdash" )
plot(
NA, type="n", xlim=c(0,30), ylim=c(0,0.8),
xlab = "Crank position (ºCA)", ylab = bquote('Burn fraction ('~X[b]~')'),
panel.first = grid()
)
with( dat, {
points( x=CA, y=Xb_exp, pch=19, col=color4["Xb_exp"], size=3 )
for( n in c("Xb_dw", "Xb_f", "Xb_s")) {
lines( x=CA, y=get(n), lty=linetypes[n], col=color4[n], lwd=2 )
}
})
legend(
x = "right",
legend = labels,
col = color4,
lty = linetypes4,
pch = c(19,NA,NA,NA),
box.lwd = 0,
inset = .02
)