本文介绍ggplt2如何给图形添加注释图层（Annotation layer）。
建议：左右上下对照来阅读，查看修改带来的变动。
续前篇：
R可视化01|ggplot2-ggplot2简介
R可视化02|ggplot2-ggplot2快速绘图
R可视化03|ggplot2图层-几何对象图层（geom layer）
R可视化04|ggplot2图层-统计变换图层（stat layer）
1、添加文本（Text labels）
2、 添加注释（Building custom annotations）
3、分面图中添加注释（Annotation across facets）
geom_text()添加文本及其在图中的横纵坐标，可修改文本字体（family）、磅值（fontface）、水平位置hjust (“left”, “center”, “right”, “inward”, “outward”)、垂直位置vjust (“bottom”, “middle”, “top”, “inward”, “outward”)、大小（size）、倾斜度（angle）、文字距原坐标点的距离（nudge）、防文本重叠（check_overlap）；防文本重叠（ggrepel::geom_text_repel）
options(repr.plot.width = 8, repr.plot.height = 10, repr.plot.res = 300) df <- data.frame(x = 1, y = 3:1, family = c("sans", "serif", "mono")) #文本字体family aesthetic p1 <- ggplot(df, aes(x, y)) + geom_text(aes(label = family, family = family)) #文本磅值（fontface） df <- data.frame(x = 1, y = 3:1, face = c("plain", "bold", "italic")) p2 <- ggplot(df, aes(x, y)) + geom_text(aes(label = face, fontface = face)) #文本水平位置hjust (“left”, “center”, “right”, “inward”, “outward”)、 #垂直位置vjust (“bottom”, “middle”, “top”, “inward”, “outward”) df <- data.frame( x = c(1, 1, 2, 2, 1.5), y = c(1, 2, 1, 2, 1.5), text = c( "bottom-left", "bottom-right", "top-left", "top-right", "center" ) ) p3 <- ggplot(df, aes(x, y)) + geom_text(aes(label = text)) p4 <- ggplot(df, aes(x, y)) + geom_text(aes(label = text), vjust = "inward", hjust = "inward")#设置水平垂直位置 #文字距原坐标点的距离（nudge） df <- data.frame(trt = c("a", "b", "c"), resp = c(1.2, 3.4, 2.5)) p5 <- ggplot(df, aes(resp, trt)) + geom_point() + geom_text(aes(label = paste0("(", resp, ")")), nudge_y = -0.25) + xlim(1, 3.6) #防文本重叠（check_overlap） p6 <- ggplot(mpg, aes(displ, hwy)) + geom_text(aes(label = model)) + xlim(1, 8) p7 <- ggplot(mpg, aes(displ, hwy)) + geom_text(aes(label = model), check_overlap = TRUE) + xlim(1, 8) #防文本重叠（ggrepel::geom_text_repel） library('ggrepel') mini_mpg <- mpg[sample(nrow(mpg), 20),] p8 <- ggplot(mpg, aes(displ, hwy)) + geom_point(colour = "red") + ggrepel::geom_text_repel(data = mini_mpg, aes(label = class)) #添加文本背景框 label <- data.frame( waiting = c(55, 80), eruptions = c(2, 4.3), label = c("peak one", "peak two") ) p9 <- ggplot(faithfuld, aes(waiting, eruptions)) + geom_tile(aes(fill = density)) + geom_label(data = label, aes(label = label)) p10 <- grid.arrange(p1,p2,p3,p4,p5,p6,p7,p8,p9,nrow = 5) ggsave("plot1.png", p10, width = 8, height = 5)ggplot2 package provides several other tools to annotate plots using the same geoms you would use to display data.
geom_text()、 geom_label() 添加文本 text, 见part1。geom_rect() ：用一个矩形圈出感兴趣的区域，指定xmin，xmax，ymin，ymax。geom_line()、geom_path()、geom_segment() ：在图形中添加线条、arrow() 可以用来添加箭头，可指定angle, length, ends and type。geom_vline()、geom_hline()、geom_abline()：添加垂直线、水平线、添加任意斜率和截距地直线。directlabels::geom_dl：directlabels直接加标签。ggforce::geom_mark_ellipse：ggforce突出某部分。gghighlight::gghighlight：gghighlight突出某部分。annotate：annotate添加文本。以实例介绍各函数的使用
options(repr.plot.width = 8, repr.plot.height = 13, repr.plot.res = 300) #添加折线geom_line() p1 <- ggplot(economics, aes(date, unemploy)) + geom_line() #添加矩形区域线geom_rect()，指定xmin，xmax，ymin，ymax presidential <- subset(presidential, start > economics$date[1]) p2 <- ggplot(economics) + geom_rect( aes(xmin = start, xmax = end, fill = party), ymin = -Inf, ymax = Inf, alpha = 0.2, data = presidential ) + geom_vline(#添加垂直线geom_vline() aes(xintercept = as.numeric(start)), data = presidential, colour = "grey50", alpha = 0.5 ) + geom_text(#添加文本geom_text() aes(x = start, y = 2500, label = name), data = presidential, size = 3, vjust = 0, hjust = 0, nudge_x = 50 ) + geom_line(aes(date, unemploy)) + scale_fill_manual(values = c("blue", "red")) + xlab("date") + ylab("unemployment") #geom_text左上侧添加文本方法一【略显复杂】 yrng <- range(economics$unemploy) xrng <- range(economics$date) caption <- paste(strwrap("Unemployment rates in the US have varied a lot over the years", 40), collapse = "\n") p3 <- ggplot(economics, aes(date, unemploy)) + geom_line() + geom_text( aes(x, y, label = caption), data = data.frame(x = xrng[1], y = yrng[2], caption = caption), hjust = 0, vjust = 1, size = 4 ) #annotate左上侧添加文本方法二【推荐】 p4 <- ggplot(economics, aes(date, unemploy)) + geom_line() + annotate( geom = "text", x = xrng[1], y = yrng[2], label = caption, hjust = 0, vjust = 1, size = 4 ) ##给某一类点上色 library('dplyr') p5 <- ggplot(mpg, aes(displ, hwy)) + geom_point(data = filter(mpg, manufacturer == "subaru"), colour = "orange", size = 3 ) + geom_point() #右上角添加图例 p6 <- p5 + annotate(geom = "point", x = 5.5, y = 40, colour = "orange", size = 3) + annotate(geom = "point", x = 5.5, y = 40) + annotate(geom = "text", x = 5.6, y = 40, label = "subaru", hjust = "left") p7 <- p5 + annotate(#添加箭头 geom = "curve", x = 4, y = 35, xend = 2.65, yend = 27, curvature = .3, arrow = arrow(length = unit(2, "mm")) ) + annotate(geom = "text", x = 4.1, y = 35, label = "subaru", hjust = "left")#添加文本 #按class添加颜色 p8 <- ggplot(mpg, aes(displ, hwy, colour = class)) + geom_point() #directlabels::geom_dl直接加标签 library("directlabels") p9 <- ggplot(mpg, aes(displ, hwy, colour = class)) + geom_point(show.legend = FALSE) + directlabels::geom_dl(aes(label = class), method = "smart.grid") #ggforce扩展包突出某部分 library('ggforce') p9 <- ggplot(mpg, aes(displ, hwy)) + geom_point() + ggforce::geom_mark_ellipse(aes(label = cyl, group = cyl)) #gghighlight扩展包突出某部分 library('gghighlight') data(Oxboys, package = "nlme") p10 <- ggplot(Oxboys, aes(age, height, group = Subject)) + geom_line() + geom_point() + gghighlight::gghighlight(Subject %in% 1:3) grid.arrange(p1,p2,p3,p4,p5,p6,p7,p8,p9,p10,nrow = 5)直接看代码。
options(repr.plot.width = 5, repr.plot.height = 7, repr.plot.res = 300) p1 <- ggplot(diamonds, aes(log10(carat), log10(price))) + geom_bin2d() + facet_wrap(vars(cut), nrow = 1) #分面图添加参考线 mod_coef <- coef(lm(log10(price) ~ log10(carat), data = diamonds)) p2 <- ggplot(diamonds, aes(log10(carat), log10(price))) + geom_bin2d() + geom_abline(intercept = mod_coef[1], slope = mod_coef[2], colour = "white", size = 1) + facet_wrap(vars(cut), nrow = 1) #gghighlight::gghighlight高亮某类 p3 <- ggplot(mpg, aes(displ, hwy, colour = factor(cyl))) + geom_point() + gghighlight::gghighlight() + facet_wrap(vars(cyl)) grid.arrange(p1,p2,p3,nrow = 3)参考资料：
ggplot2: Elegant Graphics for Data Analysis
本文结束，更多好文，欢迎关注：pythonic生物人
Python可视化|Matplotlib39-Matplotlib 1.4W+字教程（珍藏版）Python可视化|Matplotlib&Seaborn36（完结篇）
python3基础12详解模块和包（库）|构建|使用
Perl基础系列合集
NGS各种组学建库原理（图解）
