我可以在 R 中的 ggraph/ggplot2 的弧图中分隔两组顶点吗?

问题描述

我使用以下代码通过两个不同的电极得到了一个很棒的图表:

elec <- c("Fp1","Fp2","F7","F3","Fz","F4","F8","FC5","FC1","FC2","FC6","C3","Cz","C4","CP5","CP1","CP2","CP6","P7","P3","Pz","P4","P8","POz","Oz","Fp1","Oz")

edgelist <- get.edgelist(net)
# get vertex labels
label <- get.vertex.attribute(net,"name")
# get vertex groups
group <- get.vertex.attribute(net,"group")
# get vertex fill color
fill <- get.vertex.attribute(net,"color")
# get family
family <- get.vertex.attribute(net,"family")
# get vertex degree
degrees <- degree(net)

# data frame with groups,degree,labels and id
nodes <- data.frame(group,degrees,family,label,fill,id=1:vcount(net))
nodes$family <- factor(nodes$family,levels = unique(nodes$family))
nodes$label <- factor(nodes$label,levels = unique(nodes$label))
nodes <- as_tibble(nodes)

# prepare data for edges
edges <- as_tibble(edgelist)

net.tidy <- tbl_graph(nodes = nodes,edges = edges,directed = TRUE,node_key = "label")

ggraph(net.tidy,layout = "linear") + 
  geom_edge_arc(alpha = 0.5) + 
  scale_edge_width(range = c(0.2,2)) +
  scale_colour_manual(values= vrtxc) +
  geom_node_point(aes(size = degrees,color = family)) +
  geom_node_text(aes(label = elec),angle = 90,hjust = 1,nudge_y = -0.5,size = 3) +   
  coord_cartesian(clip = "off") + 
  theme_graph()+
  theme(legend.position = "top") 

我得到了一张我喜欢的很棒的图表。

Graph

但是,我想将两组电极分开,在中间,Oz 所在的位置,一点点,看看有什么不同。在节点属性中,我将它们按组 (1,2) 进行区分,我想知道是否可以使用此信息来扩展 x 轴上的两组顶点,左侧一组 25 个电极,另一组一个在轴的右边,在它们中间留一个空格。

我附上一些数据,以防它有用

> nodes
# A tibble: 50 x 6
   group degrees family label fill       id
   <fct>   <dbl> <fct>  <fct> <fct>   <int>
 1 1           5 fronp  Fp1_1 #3B9AB2     1
 2 1           9 fronp  Fp2_1 #3B9AB2     2
 3 1           6 fron   F7_1  #5DAABC     3
 4 1           7 fron   F3_1  #5DAABC     4
 5 1          11 fron   Fz_1  #5DAABC     5
 6 1           9 fron   F4_1  #5DAABC     6
 7 1          11 fron   F8_1  #5DAABC     7
 8 1           8 fronc  FC5_1 #88BAAE     8
 9 1           6 fronc  FC1_1 #88BAAE     9
10 1           4 fronc  FC2_1 #88BAAE    10
# … with 40 more rows

解决方法

如果有人感兴趣,我最终在 nodes 数据框中插入了一些虚拟行,结果成功了。