使用 ggplot2 和 directlabels 标记点
Labelling points with ggplot2 and directlabels
这是我 的后续。虽然 krlmlr 的回答以某种方式帮助解决了我的问题，但仍然存在一个问题 - 标签远离这些点，以至于无法看到哪个标签对应于哪个点。有谁知道如何解决这个问题？
测试数据
test <- structure(list(ID = c(183, 184, 185, 186, 187, 188, 189, 190,
191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 202, 203, 204
), group = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 10L, 1L, 1L, 11L, 1L, 10L, 10L, 1L, 1L, 1L), .Label = c("a",
"b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n",
"o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z"), class = "factor"),
x = c(27.4921834914348, 25.9627681619548, 30.4138361188149,
29.7795838507818, 32.33351964819, 31.9669266268744, 35.4433172141507,
37.8161067968601, 43.1590654001956, 44.2747819266045, 46.3829222044342,
42.2074195880057, 45.4532493462957, 48.393281430824, 51.7145681408198,
51.4911582677409, 51.9582538485293, 51.3328966791848, 36.4638478746633,
36.6113033420253, 39.3476493044906), y = c(-40.6667236544384,
-39.2640436250506, -40.6403794753022, -40.907139119954, -39.9691483441288,
-39.7485517513382, -38.595662907188, -38.2106224386729, -37.6418188651769,
-32.7096448721895, -34.1161006958616, -32.2821584775403,
-30.8436917254975, -30.3865899667262, -30.3910690518699,
-26.1013343566452, -23.8437232732877, -21.5548787351057,
5.50922747751602, 5.64434551903915, 5.01263995541617)), row.names = c(NA,
-21L), .Names = c("ID", "group", "x", "y"), class = "data.frame")
library(ggplot2)
library(directlabels)
绘图使用 geom_text - 问题：标签重叠
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group)) +
geom_text(aes(label=ID), show_guide=F)
按照 的建议绘图 - 问题：标签离点很远
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group)) +
geom_dl(aes(label = ID), method = defaultpf.ggplot("point",,,))
也许 hjust 和 vjust 就是您要找的东西？
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group)) +
geom_text(aes(label=ID), show_guide=F, hjust = 1.2, vjust = 0.5)
我想你可以通过调整抖动参数来做到这一点 -- ... position = position_jitter...。由于您只向我们提供了 10% 的数据，您可能需要稍微尝试一下：
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group), position= position_jitter(width= 1.5, height= 1)) +
geom_text(aes(label=ID), show_guide=F, hjust = 1.2, vjust = -.5,
position= position_jitter(width= 1.5, height= 1))
避免重叠（至少在某种程度上）的一种方法是将每个标签偏移一个由离它最近的点确定的量。因此，例如，如果一个点的最近邻点直接位于其右侧，则其标签将放置在左侧等。
# centre and normalise variables
test$yy <- (test$y - min(test$y)) / (max(test$y) - min(test$y))
test$xx <- (test$x - min(test$x)) / (max(test$x) - min(test$x))
test$angle <- NA
for (i in 1:nrow(test)) {
dx <- test[-i, ]$xx - test[i, ]$xx
dy <- test[-i, ]$yy - test[i, ]$yy
j <- which.min(dx ^ 2 + dy ^ 2)
theta <- atan2((test[-i, ]$yy[j] - test[i, ]$yy), (test[-i, ]$xx[j] - test[i, ]$xx))
test[i, ]$angle <- theta + pi
}
sc <- 0.5
test$nudge.x <- cos(test$angle) * sc
test$nudge.y <- sin(test$angle) * sc
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group)) +
geom_text(aes(x = x + nudge.x, y = y + nudge.y, label = ID), size = 3, show.legend = FALSE)
您可以尝试调整缩放参数 sc（它越大，标签离点越远）以避免标签重叠。 （我想可能会发生不同的 sc 可以应用于所有点以避免重叠 - 在这种情况下，您需要更改每个点的缩放参数，可能通过使用 [=14] 定义 sc =] 和 dy)。
可能 ggrepel 更适合标记散点图中的点。
library(ggplot2) # ggrepel requires ggpot2 v2.0.0
library(ggrepel)
ggplot(test, aes(x=x, y=y)) +
geom_text_repel(aes(label = ID, color = group), show.legend = FALSE,
box.padding = unit(0.45, "lines")) +
geom_point(aes(colour=group))
这是我
测试数据
test <- structure(list(ID = c(183, 184, 185, 186, 187, 188, 189, 190,
191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 202, 203, 204
), group = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 10L, 1L, 1L, 11L, 1L, 10L, 10L, 1L, 1L, 1L), .Label = c("a",
"b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n",
"o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z"), class = "factor"),
x = c(27.4921834914348, 25.9627681619548, 30.4138361188149,
29.7795838507818, 32.33351964819, 31.9669266268744, 35.4433172141507,
37.8161067968601, 43.1590654001956, 44.2747819266045, 46.3829222044342,
42.2074195880057, 45.4532493462957, 48.393281430824, 51.7145681408198,
51.4911582677409, 51.9582538485293, 51.3328966791848, 36.4638478746633,
36.6113033420253, 39.3476493044906), y = c(-40.6667236544384,
-39.2640436250506, -40.6403794753022, -40.907139119954, -39.9691483441288,
-39.7485517513382, -38.595662907188, -38.2106224386729, -37.6418188651769,
-32.7096448721895, -34.1161006958616, -32.2821584775403,
-30.8436917254975, -30.3865899667262, -30.3910690518699,
-26.1013343566452, -23.8437232732877, -21.5548787351057,
5.50922747751602, 5.64434551903915, 5.01263995541617)), row.names = c(NA,
-21L), .Names = c("ID", "group", "x", "y"), class = "data.frame")
library(ggplot2)
library(directlabels)
绘图使用 geom_text - 问题：标签重叠
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group)) +
geom_text(aes(label=ID), show_guide=F)
按照 的建议绘图 - 问题：标签离点很远
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group)) +
geom_dl(aes(label = ID), method = defaultpf.ggplot("point",,,))
也许 hjust 和 vjust 就是您要找的东西？
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group)) +
geom_text(aes(label=ID), show_guide=F, hjust = 1.2, vjust = 0.5)
我想你可以通过调整抖动参数来做到这一点 -- ... position = position_jitter...。由于您只向我们提供了 10% 的数据，您可能需要稍微尝试一下：
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group), position= position_jitter(width= 1.5, height= 1)) +
geom_text(aes(label=ID), show_guide=F, hjust = 1.2, vjust = -.5,
position= position_jitter(width= 1.5, height= 1))
避免重叠（至少在某种程度上）的一种方法是将每个标签偏移一个由离它最近的点确定的量。因此，例如，如果一个点的最近邻点直接位于其右侧，则其标签将放置在左侧等。
# centre and normalise variables
test$yy <- (test$y - min(test$y)) / (max(test$y) - min(test$y))
test$xx <- (test$x - min(test$x)) / (max(test$x) - min(test$x))
test$angle <- NA
for (i in 1:nrow(test)) {
dx <- test[-i, ]$xx - test[i, ]$xx
dy <- test[-i, ]$yy - test[i, ]$yy
j <- which.min(dx ^ 2 + dy ^ 2)
theta <- atan2((test[-i, ]$yy[j] - test[i, ]$yy), (test[-i, ]$xx[j] - test[i, ]$xx))
test[i, ]$angle <- theta + pi
}
sc <- 0.5
test$nudge.x <- cos(test$angle) * sc
test$nudge.y <- sin(test$angle) * sc
ggplot(test, aes(x=x, y=y)) +
geom_point(aes(colour=group)) +
geom_text(aes(x = x + nudge.x, y = y + nudge.y, label = ID), size = 3, show.legend = FALSE)
您可以尝试调整缩放参数 sc（它越大，标签离点越远）以避免标签重叠。 （我想可能会发生不同的 sc 可以应用于所有点以避免重叠 - 在这种情况下，您需要更改每个点的缩放参数，可能通过使用 [=14] 定义 sc =] 和 dy)。
可能 ggrepel 更适合标记散点图中的点。
library(ggplot2) # ggrepel requires ggpot2 v2.0.0
library(ggrepel)
ggplot(test, aes(x=x, y=y)) +
geom_text_repel(aes(label = ID, color = group), show.legend = FALSE,
box.padding = unit(0.45, "lines")) +
geom_point(aes(colour=group))