我在使用 R 中提供的 igraph 包时遇到了有关网络可视化的问题。
假设您有一个特定的网络，其中包含节点和边的总样本。让我将这个网络命名为 netX:
netX <- structure(c(1, 0.48275862, 0.51724138, 0.48275862, 0.27906977,
0.06896552, 0.34482759, 0.32352941, 0.06896552, 0.34482759, 0.03448276,
0.06896552, 0.20689655, 0.17241379, 0.17241379, 0, 0.23333333,
0.27586207, 0.21621622, 0.24137931, 0.48275862, 1, 0.4137931,
0.35714286, 0.25581395, 0.25, 0.25, 0.38235294, 0.07142857, 0.28571429,
0.21428571, 0.28571429, 0.10714286, 0.07142857, 0.21428571, 0.03571429,
0.2, 0.32142857, 0.16216216, 0.25, 0.51724138, 0.4137931, 1,
0.5862069, 0.34883721, 0.06896552, 0.20689655, 0.32352941, 0,
0.27586207, 0.10344828, 0.06896552, 0.10344828, 0.10344828, 0.13793103,
0, 0.36666667, 0.24137931, 0.21621622, 0.17241379, 0.48275862,
0.35714286, 0.5862069, 1, 0.23255814, 0.11538462, 0.23076923,
0.26470588, 0, 0.30769231, 0.11538462, 0.07692308, 0.11538462,
0.15384615, 0.25, 0.03846154, 0.26666667, 0.21428571, 0.21621622,
0.14285714, 0.27906977, 0.25581395, 0.34883721, 0.23255814, 1,
0.18604651, 0.11627907, 0.18604651, 0, 0.18604651, 0.25581395,
0.25581395, 0.27906977, 0.13953488, 0.20930233, 0.04651163, 0.23255814,
0.34883721, 0.25581395, 0.30232558, 0.06896552, 0.25, 0.06896552,
0.11538462, 0.18604651, 1, 0.13043478, 0.32352941, 0, 0.25, 0.34782609,
0.34782609, 0.16666667, 0.13043478, 0.35714286, 0.17391304, 0.16666667,
0.28571429, 0.16216216, 0.35714286, 0.34482759, 0.25, 0.20689655,
0.23076923, 0.11627907, 0.13043478, 1, 0.23529412, 0, 0.29166667,
0.04347826, 0.09090909, 0.16666667, 0.33333333, 0.21428571, 0.06666667,
0.23333333, 0.17857143, 0.16216216, 0.21428571, 0.32352941, 0.38235294,
0.32352941, 0.26470588, 0.18604651, 0.32352941, 0.23529412, 1,
0.08823529, 0.29411765, 0.23529412, 0.17647059, 0.11764706, 0.14705882,
0.26470588, 0.08823529, 0.26470588, 0.26470588, 0.32432432, 0.23529412,
0.06896552, 0.07142857, 0, 0, 0, 0, 0, 0.08823529, 1, 0.04166667,
0, 0, 0, 0, 0.03571429, 0, 0, 0, 0.05405405, 0.03571429, 0.34482759,
0.28571429, 0.27586207, 0.30769231, 0.18604651, 0.25, 0.29166667,
0.29411765, 0.04166667, 1, 0.25, 0.25, 0.20833333, 0.125, 0.25,
0.04166667, 0.23333333, 0.17857143, 0.16216216, 0.14285714, 0.03448276,
0.21428571, 0.10344828, 0.11538462, 0.25581395, 0.34782609, 0.04347826,
0.23529412, 0, 0.25, 1, 0.69565217, 0.125, 0.08695652, 0.17857143,
0, 0.16666667, 0.32142857, 0.24324324, 0.25, 0.06896552, 0.28571429,
0.06896552, 0.07692308, 0.25581395, 0.34782609, 0.09090909, 0.17647059,
0, 0.25, 0.69565217, 1, 0.08333333, 0.09090909, 0.17857143, 0,
0.2, 0.35714286, 0.18918919, 0.25, 0.20689655, 0.10714286, 0.10344828,
0.11538462, 0.27906977, 0.16666667, 0.16666667, 0.11764706, 0,
0.20833333, 0.125, 0.08333333, 1, 0.20833333, 0.25, 0.04166667,
0.2, 0.17857143, 0.27027027, 0.28571429, 0.17241379, 0.07142857,
0.10344828, 0.15384615, 0.13953488, 0.13043478, 0.33333333, 0.14705882,
0, 0.125, 0.08695652, 0.09090909, 0.20833333, 1, 0.21428571,
0.15384615, 0.2, 0.25, 0.16216216, 0.14285714, 0.17241379, 0.21428571,
0.13793103, 0.25, 0.20930233, 0.35714286, 0.21428571, 0.26470588,
0.03571429, 0.25, 0.17857143, 0.17857143, 0.25, 0.21428571, 1,
0.10714286, 0.26666667, 0.32142857, 0.37837838, 0.46428571, 0,
0.03571429, 0, 0.03846154, 0.04651163, 0.17391304, 0.06666667,
0.08823529, 0, 0.04166667, 0, 0, 0.04166667, 0.15384615, 0.10714286,
1, 0.16666667, 0.10714286, 0.05405405, 0.07142857, 0.2333333,
0.2, 0.3666667, 0.2666667, 0.2325581, 0.1666667, 0.2333333, 0.2647059,
0, 0.2333333, 0.1666667, 0.2, 0.2, 0.2, 0.2666667, 0.1666667,
1, 0.4, 0.2972973, 0.3, 0.2758621, 0.3214286, 0.2413793, 0.2142857,
0.3488372, 0.2857143, 0.1785714, 0.2647059, 0, 0.1785714, 0.3214286,
0.3571429, 0.1785714, 0.25, 0.3214286, 0.1071429, 0.4, 1, 0.1891892,
0.4285714, 0.21621622, 0.16216216, 0.21621622, 0.21621622, 0.25581395,
0.16216216, 0.16216216, 0.32432432, 0.05405405, 0.16216216, 0.24324324,
0.18918919, 0.27027027, 0.16216216, 0.37837838, 0.05405405, 0.2972973,
0.18918919, 1, 0.43243243, 0.24137931, 0.25, 0.17241379, 0.14285714,
0.30232558, 0.35714286, 0.21428571, 0.23529412, 0.03571429, 0.14285714,
0.25, 0.25, 0.28571429, 0.14285714, 0.46428571, 0.07142857, 0.3,
0.42857143, 0.43243243, 1), .Dim = c(20L, 20L), .Dimnames = list(
c("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k",
"l", "m", "n", "o", "p", "q", "r", "s", "t"), c("a", "b",
"c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n",
"o", "p", "q", "r", "s", "t")))
进一步假设我有这个网络的某个子集，我称之为 netY:
sub_netX <- c("a", "d", "h", "l", "o")
netY <- netX[sub_netX,sub_netX]
netY
a d h l o
a 1.00000000 0.48275862 0.3235294 0.06896552 0.1724138
d 0.48275862 1.00000000 0.2647059 0.07692308 0.2500000
h 0.32352941 0.26470588 1.0000000 0.17647059 0.2647059
l 0.06896552 0.07692308 0.1764706 1.00000000 0.1785714
o 0.17241379 0.25000000 0.2647059 0.17857143 1.0000000
我的目标是覆盖两个网络，以便可以在后台看到完整的网络 netX，在前台看到子集 netY。但首先我们需要加载 igraph 包，确定势图的布局并存储 X 和 Y 坐标。
library(igraph)
inetX <- graph.adjacency(netX, mode = "undirected", weighted = TRUE, diag=F)
inetY <- graph.adjacency(netY, mode = "undirected", weighted = TRUE, diag=F)
lay <- layout.fruchterman.reingold(inetX)
V(inetX)$x <- lay[, 1]
V(inetX)$y <- lay[, 2]
此外，我认为获取子集的索引并不是真正必要的，但它可以方便以后进行索引:
idx <- which(V(inetX)$name %in% c("a", "d", "h", "l", "o"))
我们还可以预先存储节点颜色:
V(inetX)$color <- "grey"
V(inetY)$color <- "red"
但是，当我尝试在一个简单的图中覆盖两个网络时发生的情况是，尽管坐标固定，但子集 inetY 的节点并未完全显示在 inetX 的相同节点的位置。
plot.igraph(inetX, layout = lay, vertex.size = 7)
plot.igraph(inetY, layout = lay[idx,], vertex.size = 4, add = TRUE)
由于我的声誉太低，我无法上传图片，但我希望您可以按照您机器上的步骤进行操作。
我对你们所有人的问题是，为什么会发生这种情况？在覆盖两个网络时如何修复节点坐标的意义上，我做错了什么？
非常感谢您提前!
最佳答案
原因是 igraph 绘图函数将布局重新缩放为 [-1,1] x [-1,1]默认情况下。您可以使用 rescale=FALSE论据 plot()然后不执行重新缩放。
请注意，在这种情况下，您需要通过设置 xlim 手动设置绘图的限制。和 ylim :
xlim <- range(lay[,1])
ylim <- range(lay[,2])
plot.igraph(inetX, layout = lay, vertex.size = 20,
xlim = xlim, ylim = ylim, rescale = FALSE)
plot.igraph(inetY, layout = lay[idx,], vertex.size = 14,
add = TRUE, rescale = FALSE)
关于r - 使用 igraph R 在完整网络上绘制子集时如何修复节点，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/24301992/