使用R语言igraph包绘制带地理坐标节点时位置错误的问题排查

Let's break down the issues in your code and fix them to get the correct geographic positioning of your nodes:

1. You're referencing the wrong vertex property for matching #

When you create a graph with graph_from_data_frame(), the vertex names are stored in V(main)$name—not V(main)$from. The from field only exists in your original edge data frame, not as a vertex attribute. Your original code tried to match V(main)$from (which doesn't exist) to meta$name, leading to incorrect (or possibly NA) latitude/longitude values assigned to your vertices.

2. Your layout matrix has reversed coordinates #

In standard geographic plotting, the x-axis corresponds to longitude (east-west position) and the y-axis corresponds to latitude (north-south position). You had latitude first in your matrix, which swaps the axes and puts nodes in the wrong places.

Fixed Full Code #

library(igraph)

# Create edge data and graph
city <- data.frame(from = c("London", "Paris", "Beijing", "Berlin"), to = c("Beijing", "Berlin", "London", "Paris"))
main <- graph_from_data_frame(city)

# Metadata
meta <- data.frame(name = c("Paris", "Hong Kong", "London", "Rome", "Berlin", "Beijing", "Madrid"), 
                   latitude = c(48.52, 22.26, 51.30, 41.54, 52.30, 39.9, 40.23), 
                   longitude = c(2.17, 114.12, 0.10, 12.29, 13.25, 116.3, 3.43))

# Correctly assign vertex attributes using vertex names
V(main)$latitude <- meta$latitude[match(V(main)$name, meta$name)]
V(main)$longitude <- meta$longitude[match(V(main)$name, meta$name)]

# Build layout matrix with correct coordinate order: longitude (x) first, latitude (y) second
location <- matrix(c(V(main)$longitude, V(main)$latitude), ncol = 2)

# Plot with labels to verify positions
plot(main, vertex.label = V(main)$name, vertex.size = 8, edge.width = 1, layout = location)

What This Fixes #

After running this code, you'll see:

• Beijing (high longitude, mid latitude) on the far right
• London (low longitude, high latitude) on the far left
• Paris and Berlin clustered in the middle-left, matching their actual European locations

This should resolve the incorrect positioning issue you were seeing.

内容的提问来源于stack exchange，提问作者Yang Yuan