use a community matrix do spectral analysis using k means, and plot a cluster graph with ordered centers in r

library(vegan)
library(factoextra)

# Load your community matrix (replace "your_matrix.csv" with the actual path to your matrix)
data <- read.csv("your_matrix.csv", row.names = 1)
community_matrix <- as.matrix(data)

# Perform spectral analysis
community_pca <- rda(community_matrix)
eig_vals <- eigenvals(community_pca)
sort_eig <- rev(sort(eig_vals, decreasing = TRUE))

# Determine optimal number of clusters using 'factoextra' package
fviz_nbclust(community_matrix, kmeans, method = "wss")

# Perform k-means clustering
num_clusters <- 3 # Choose the optimal number of clusters based on the plot
kmeans_result <- kmeans(community_matrix, centers = num_clusters)

# Plot cluster graph with ordered centers
fviz_cluster(kmeans_result, geom = "point", data = community_matrix, stand = F, label = "none")
805 chars
22 lines