R语言circlize包：弦图highlight.sector()宽度与位置调整求助

Got it, let's work through how to fine-tune the highlight.sector() function in circlize for your fishery landing data. I’ll walk you through practical examples tailored to your setup—where you’ve got home ports (PORT_DE), landing ports (PORT_LA), scallop weights (SCALLOP_W), and port metadata tied to regions (Region_lb) and countries (Country_lb).

First, let’s start with a quick setup using simulated data that mirrors your structure, so you can adapt the code directly to your real dataset:

library(circlize)
library(reshape2)

# Simulate your fishery data
set.seed(123)
port_data <- data.frame(
  PORT_DE = sample(c("PortA_FR", "PortB_FR", "PortC_UK", "PortD_UK"), 100, replace = TRUE),
  PORT_LA = sample(c("PortX_FR", "PortY_UK", "PortZ_ES"), 100, replace = TRUE),
  SCALLOP_W = rnorm(100, mean = 5, sd = 2)
)

# Port metadata (match your Name_short, Region_lb, Country_lb fields)
port_meta <- data.frame(
  Name_short = c("PortA", "PortB", "PortC", "PortD", "PortX", "PortY", "PortZ"),
  Region_lb = c("Brittany", "Brittany", "Cornwall", "Cornwall", "Normandy", "Scotland", "Galicia"),
  Country_lb = c("France", "France", "UK", "UK", "France", "UK", "Spain")
)

# Build adjacency matrix (sum scallop weights by port pair)
adj_matrix <- aggregate(SCALLOP_W ~ PORT_DE + PORT_LA, data = port_data, sum)
adj_matrix <- dcast(adj_matrix, PORT_DE ~ PORT_LA, value.var = "SCALLOP_W", fill = 0)
rownames(adj_matrix) <- adj_matrix$PORT_DE
adj_matrix <- adj_matrix[, -1]

# Base chord diagram with annotation track
chordDiagram(adj_matrix, 
             annotationTrack = "grid", 
             preAllocateTracks = list(track.height = 0.1))
circos.trackPlotRegion(track.index = 1, panel.fun = function(x, y) {
  circos.text(CELL_META$xcenter, CELL_META$ylim[1], CELL_META$sector.index, 
              facing = "clockwise", niceFacing = TRUE, adj = c(0, 0.5))
}, bg.border = NA)

1. Adjust Width & Position for Single Sectors #

To tweak the highlight size and placement for individual ports (e.g., a key home port), use the padding parameter—it controls the inner margin of the highlight relative to the full sector width (values are proportional to the sector’s angle, 0–1).

For example, to highlight PortA_FR with a narrower, centered highlight:

highlight.sector(sector.index = "PortA_FR", 
                 track.index = 1, 
                 col = "#FFDDCC", 
                 padding = c(0.1, 0, 0.1, 0), # Top/bottom margins shrink the highlight width by 20%
                 text = "French Home Port", 
                 text.vjust = -0.5, # Move text above the highlight
                 text.hjust = 0.5,
                 cex = 0.8)

• Smaller padding values = wider highlight; larger values = narrower highlight.
• Use text.vjust/text.hjust to nudge your label exactly where you want it (positive values move text down/right, negative up/left).

2. Highlight Groups of Ports (By Region/Country) #

Since your ports are grouped by Region_lb or Country_lb, you can highlight all ports in a single group and adjust the spacing between them with extend (fills gaps between adjacent sectors in the group):

# Get all French ports present in your adjacency matrix
french_ports <- c(paste0(port_meta$Name_short[port_meta$Country_lb == "France"], "_FR"),
                  paste0(port_meta$Name_short[port_meta$Country_lb == "France"], "")) # Match your port naming
french_ports <- intersect(french_ports, c(rownames(adj_matrix), colnames(adj_matrix)))

# Highlight all French ports with a unified, gap-free highlight
highlight.sector(sector.index = french_ports, 
                 track.index = 1, 
                 col = "#CCEEFF", 
                 padding = c(0.05, 0.05, 0.05, 0.05), # Small margins around the entire group
                 extend = 0.02, # Extend highlights 2% of the circle to fill gaps between sectors
                 text = "French Ports", 
                 text.vjust = -1, # Move label further outward
                 text.col = "darkblue",
                 cex = 1)

3. Precise Position Control with Angle Ranges #

For ultra-specific placement, use xlim to manually define the start/end angles of the highlight (in radians). First, grab the sector’s angle range, then adjust it:

# Get angle range for PortC_UK
portc_angle <- get.cell.meta.data("xlim", sector.index = "PortC_UK", track.index = 1)

# Shrink the highlight to the middle 80% of the sector
highlight.sector(sector.numeric.index = which(get.all.sector.index() == "PortC_UK"), 
                 track.index = 1, 
                 col = "#DDFFCC", 
                 xlim = c(portc_angle[1] + 0.1, portc_angle[2] - 0.1), # Narrow the highlight by 0.1 radians on each side
                 text = "UK Landing Port", 
                 text.vjust = 1.5, # Move label inside the sector
                 text.hjust = 0.5)

4. Adjust Highlight Position Across Tracks #

If you’ve pre-allocated multiple tracks (with preAllocateTracks), use track.index and ylim to control where the highlight sits vertically in the track:

# Add a second track first (uncomment if you didn't include it earlier)
# circos.trackPlotRegion(track.index = 2, bg.border = NA)

# Highlight PortZ_ES in the second track, centered vertically
highlight.sector(sector.index = "PortZ_ES", 
                 track.index = 2, 
                 col = "#FFCCEE", 
                 ylim = c(0.2, 0.8), # Restrict highlight to the middle 60% of the track height
                 text = "Spanish Port", 
                 text.vjust = 0.5,
                 text.hjust = 0.5)

Key Parameters Recap #

• padding: Controls highlight width relative to the sector
• extend: Connects highlights across grouped sectors
• xlim: Manually sets highlight start/end angles for precision
• ylim: Adjusts vertical position in multi-track diagrams
• text.vjust/text.hjust: Fine-tune label placement

内容的提问来源于stack exchange，提问作者Andronikos K.