Mercurial > repos > ecology > claraguess

comparison claraguess.R @ 0:52d4151e00d8 draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/galaxyecology/tools-ecology/tree/master/tools/Ecoregionalization_workflow commit ced658540f05bb07e1e687af30a3fa4ea8e4803c
author ecology
date Wed, 28 May 2025 10:12:06 +0000
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:52d4151e00d8
1 ##30/04/2025
2 ##Jean Le Cras
3 ### Clustering with Clara algorithm with an option to determine the optimal number of clusters based on SIH index
4
5 #load libraries
6 library(cluster)
7 library(dplyr)
8 library(tidyverse)
9
10 #load arguments
11 args <- commandArgs(trailingOnly = TRUE)
12 if (length(args)==0) {
13 stop("This tool needs at least one argument")
14 }
15
16 #load data
17 enviro_path <- args[1]
18 preds_path <- args[2]
19 taxas_path <- args[3]
20 type <- args[4]
21 k <- as.integer(args[5])
22 metric <- args[6]
23 samples <- as.integer(args[7])
24 env.data <- read.table(enviro_path, sep = "\t", header = TRUE, dec = ".", na.strings = "-9999")
25
26 data_split = str_split(preds_path, ",")
27 preds.data = NULL
28
29 for (i in 1:length(data_split[[1]])) {
30 df <- read.table(data_split[[1]][i], dec=".", sep="\t", header=T, na.strings="NA")
31 preds.data <- rbind(preds.data, df)
32 remove(df)
33 }
34
35 names(preds.data) <- c("lat", "long", "pred", "taxa")
36
37 development_traits <- str_split(readLines(taxas_path), "\t")
38
39 #select the clara model with the optimal number of clusters
40 model <- NULL
41
42 if (type == "auto") {
43 sih_scores <- c()
44 models <- list()
45
46 for (i in 2:k) {
47 models[[i]] <- clara(preds.data$pred, i, metric = metric, samples = samples, stand = TRUE)
48 sih_scores[i] <- models[[i]]$silinfo$avg.width
49 }
50 png("sih_scores.png")
51 plot(2:k, sih_scores[2:k], type = "b", xlab = "Number of clusters", ylab = "SIH index")
52 dev.off()
53
54 best_k <- which.max(sih_scores[3:k]) + 2
55 model <- models[[best_k]]
56 k <- best_k
57 } else {
58 model <- clara(preds.data$pred, k, metric = metric, samples = samples, stand = TRUE)
59 }
60
61 #saving results
62 png("silhouette_plot.png")
63 plot(silhouette(model), main = paste("Silhouette plot for k =", k))
64 dev.off()
65
66 data.test <- matrix(preds.data$pred, nrow = nrow(env.data), ncol = nrow(preds.data) / nrow(env.data))
67 data.test <- data.frame(data.test)
68 names(data.test) <- unique(preds.data$development)
69
70 full.data <- cbind(preds.data[1:nrow(data.test), 1:2], model$clustering)
71 names(full.data) <- c("lat", "long", "cluster")
72 full.data <- cbind(full.data, data.test, env.data[, 3:ncol(env.data)])
73
74 write.table(full.data[1:3], file = "data_cluster.tabular", quote = FALSE, sep = "\t", row.names = FALSE)
75 write.table(full.data, file = "clustered_taxas_env.tabular", quote = FALSE, sep = "\t", row.names = FALSE)