论文

Changes in plant inputs alter soil carbon and microbial communities in forest

本地pdf Global Change Biology - 2022 - Feng - Changes in plant inputs alter soil carbon and microbial communities in forest.pdf

今天的推文重复一下论文中的Figure3，这个是之前有读者在公众号后台的留言，之前我不知道怎么实现这种好几个子图中间没有空白的形式，有读者留言可以用分面然后调节主题里的参数panel.spacing = unit(0,'lines')，有了基本思路就可以尝试一下这个图

image.png

今天推文的主要内容是实现这个图的主题部分，四周的一些注释，分面图在指定的子图添加文本等等再单独的一篇推文来介绍

添加注释会用到一个自定义函数

annotation_custom2 <- function (grob, 
                                xmin = -Inf, 
                                xmax = Inf, 
                                ymin = -Inf, 
                                ymax = Inf, 
                                data) 
{
  layer(data = data, 
        stat = StatIdentity, 
        position = PositionIdentity, 
        geom = ggplot2:::GeomCustomAnn,
        inherit.aes = TRUE, params = list(grob = grob, 
                                          xmin = xmin, xmax = xmax, 
                                          ymin = ymin, ymax = ymax))
}

这个是在网上找到的，具体来源想不起来了

首先是示例数据集截图

image.png

这个不是论文中提供的，是我自己随便构造的

读取数据集

library(readxl)

dat<-read_excel("data/20220804/example_df.xlsx")
dat

最基本的点和误差线

library(ggplot2)

ggplot(data=dat,aes(x=x,y=y))+
  geom_errorbarh(aes(xmin=x-sd_value,
                     xmax=x+sd_value,
                     color=group02),
                 height=0,
                 show.legend = FALSE)+
  geom_point(aes(color=group02),
             size=5,
             show.legend = TRUE)

image.png

分面加一些主题设置

ggplot(data=dat,aes(x=x,y=y))+
  geom_errorbarh(aes(xmin=x-sd_value,
                     xmax=x+sd_value,
                     color=group02),
                 height=0,
                 show.legend = FALSE)+
  geom_point(aes(color=group02),
             size=5,
             show.legend = TRUE)+
  facet_wrap(~group01)+
  theme_bw()+
  theme(panel.spacing = unit(0,'lines'),
        panel.background = element_blank(),
        axis.line = element_line(),
        panel.grid = element_blank(),
        axis.ticks.y = element_blank(),
        strip.text = element_text(hjust = 0))

image.png

添加一些虚线，坐标轴设置

ggplot(data=dat,aes(x=x,y=y))+
  geom_errorbarh(aes(xmin=x-sd_value,
                     xmax=x+sd_value,
                     color=group02),
                 height=0,
                 show.legend = FALSE)+
  geom_point(aes(color=group02),
             size=5,
             show.legend = TRUE)+
  facet_wrap(~group01)+
  theme_bw()+
  theme(panel.spacing = unit(0,'lines'),
        panel.background = element_blank(),
        axis.line = element_line(),
        panel.grid = element_blank(),
        axis.ticks.y = element_blank(),
        strip.text = element_text(hjust = 0))+
  labs(y="",x=NULL)+
  scale_x_continuous(limits = c(-30,30),
                     breaks = seq(-30,30,10))+
  scale_y_continuous(limits = c(0.5,3.5),
                     breaks = c(1,2,3),
                     labels=c("SOC concentration",
                              "SOC stock",
                              "SOC concentration"),
                     expand = expansion(mult = c(0,0)))+
  geom_vline(xintercept = 0,
             color="gray",
             lty="dashed")+
  geom_hline(yintercept = 1,
             color="gray",
             lty="dashed")+
  geom_hline(yintercept = 2,
             color="gray",
             lty="dashed")+
  geom_hline(yintercept = 3,
             color="gray",
             lty="dashed") -> p0
p0

最后是自定义颜色

p0+
  scale_color_manual(values = c(
    "A"="#b856d7",
    "B"="#55a0fb",
    "C"="#0f99b2",
    "D"="#ffa040",
    "E"="#008000"
  ),
  labels=c(
    "A"="Litter addition",
    "B"="Litter removal",
    "C"="Root removal",
    "D"="Litter effect",
    "E"="Root effect"
    ))+
  theme(legend.position = "bottom",
        legend.title = element_blank())