-
Notifications
You must be signed in to change notification settings - Fork 0
/
NMDS_and_MDS_play.R
251 lines (200 loc) · 7.98 KB
/
NMDS_and_MDS_play.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
# SET THE STAGE -----------------------------------------------------------
# set the workign directory
setwd("~/Documents/GitHub/workflow_play/")
# simple function to import data
import_data <- function(file_name)
{
data.matrix(read.table(file_name,
row.names=1,
header=TRUE,
sep="\t",
comment.char="",
quote="", check.names=FALSE))
}
# simple function to import metadata
import_metadata <- function(group_table){ #, group_column, sample_names){
metadata_matrix <- as.matrix( # Load the metadata table (same if you use one or all columns)
read.table(
file=group_table,row.names=1,header=TRUE,sep="\t",
colClasses = "character", check.names=FALSE,
comment.char = "",quote="",fill=TRUE,blank.lines.skip=FALSE
)
)
}
# NMDS --------------------------------------------------------------------
my_data <- import_data("filtered_counts.txt")
my_metadata <- import_metadata("filtered_counts.metadata.txt")
# data expected as a sample per ROW so have to rotate it
colnames(my_data)
library(matlab)
my_data <- rot90(my_data)
rownames(my_data)
# Load the required packages
library(vegan)
library(tidyverse)
# Create a sample data matrix
# data <- matrix(c(1, 2, 3, 4, 5, 6, 7, 8, 9), nrow = 3, ncol = 3)
# Compute the distance matrix
dist_matrix <- vegdist(my_data)
# method # bray by default
# Dissimilarity index, partial match to
###"manhattan", "euclidean", "canberra", "clark", "bray", "kulczynski",
###"jaccard", "gower", "altGower", "morisita", "horn", "mountford", "raup",
###"binomial", "chao", "cao", "mahalanobis", "chisq", "chord", "hellinger",
###"aitchison", or "robust.aitchison".
# Perform NMDS on the distance matrix
nmds_result <- metaMDS(dist_matrix, k=2) # use k=3 for 3d
# Print the NMDS result
nmds_result
nmds_result$points
# take a quick look
plot(x =nmds_result$points[,"MDS1"] , y =nmds_result$points[,"MDS2"])
# prep tidy data for ggplot
# make sure the data and metadata are sorted by rowname
# Sort the rows based on row names
sorted_metadata <- my_metadata[order(rownames(my_metadata)), ]
results <- cbind(nmds_result$points[,"MDS1"], nmds_result$points[,"MDS2"])
sorted_results <- results[order(rownames(results)), ]
# make the output tidy for ggplot
my_tidy_data <- cbind(sorted_results, sorted_metadata[,"env_package.data.body_site"])
colnames(my_tidy_data) <- c("MDS1","MDS2", "env_package.data.body_site" )
my_tidy_data[1:3,]
my_tidy_data <- as.data.frame(my_tidy_data)
# plot the results with ggplot
ggplot(
data = my_tidy_data,
mapping = aes(x=MDS1, y=MDS2, color=env_package.data.body_site)) +
geom_point()
# Do a 3d NMDS -- requires functions below to generate color from metadata
dist_matrix <- vegdist(my_data)
nmds_result <- metaMDS(dist_matrix, k=3) # use k=3 for 3d
results <- cbind(nmds_result$points[,"MDS1"], nmds_result$points[,"MDS2"],nmds_result$points[,"MDS3"])
sorted_results <- results[order(rownames(results)), ]
# make the output tidy for ggplot
my_tidy_data <- cbind(sorted_results, sorted_metadata[,"env_package.data.body_site"])
colnames(my_tidy_data) <- c("MDS1","MDS2","MDS3", "env_package.data.body_site" )
my_tidy_data[1:3,]
my_tidy_data <- as.data.frame(my_tidy_data)
# plot it WITHOUT GGPLOT
# Create colors from the metadata
# generate metadata colors here
metadata_column = my_tidy_data[,'env_package.data.body_site', drop=FALSE]
metadata_column <- metadata_column[order(rownames(metadata_column)),,drop=FALSE]
# generate colors for selected column
metadata_colors <- create_colors(metadata_column)
# render 3d plot
plot_ly( x=my_tidy_data$MDS1,
y=my_tidy_data$MDS2,
z=my_tidy_data$MDS3,
type = "scatter3d",
mode = "markers",
#marker = list(color="red") # list(color = metadata_colors[,1]))
marker = list(color = metadata_colors[,1])
)
# MDS ---------------------------------------------------------------------
# import data and metadata and make sure that both are similarly sorted by mgid
my_data <- import_data("filtered_counts.txt")
my_metadata <- import_metadata("filtered_counts.metadata.txt")
identical(colnames(my_data), rownames(my_metadata))
my_data <- my_data[,order(colnames(my_data))]
my_metadata <- my_metadata[order(rownames(my_metadata)),]
identical(colnames(my_data), rownames(my_metadata))
# data expected as a sample per ROW so have to rotate it
colnames(my_data)
library(matlab)
my_data <- rot90(my_data)
rownames(my_data)
colnames(my_data)
# create dist matrix
dist_matrix <- vegdist(my_data)
# Perform Multidimensional Scaling (MDS)
mds_result <- cmdscale(dist_matrix, k = 3) # k=3for 3D visualization
# Extract coordinates for plotting
mds_coordinates <- as.data.frame(mds_result)
mds_coordinates[1:3,]
# Quick Plot of the MDS result
plot(mds_coordinates$V1, mds_coordinates$V2, type = "n", xlab = "Dimension 1", ylab = "Dimension 2", main = "MDS Plot")
text(mds_coordinates$V1, mds_coordinates$V2, labels = rownames(mds_coordinates))
# make tidy and prep for scatterplot 3d and plotly
# first make sure data and metadata are sorted the same
my_data <- mds_coordinates[order(rownames(mds_coordinates)),]
my_metadata <- my_metadata[order(rownames(my_metadata)),]
# combine data and metadata to make it tidy
my_tidy_data <- cbind(my_data,my_metadata[,"env_package.data.body_site"])
colnames(my_tidy_data) <- c("V1","V2","V3", "env_package.data.body_site" )
# make a 2 d plot with ggplot
ggplot(
data = my_tidy_data,
mapping = aes(x=V1, y=V2, color=env_package.data.body_site)) +
geom_point()
# Now do an interactive 3d plot
library(plotly)
library(scatterplot3d)
# FUNCTIONS TO GENERATE COLOR FROM METADATA -------------------------------
# Functions to create colors from metadata
## ######################
## # SUB(3): FUNCTION TO GENERATE COLORS FOR THE PLOT
## ######################
create_colors <- function(metadata_column, color_mode = "auto"){ # function to
my_data.color <- data.frame(metadata_column)
column_factors <- as.factor(metadata_column[,1])
column_levels <- levels(as.factor(metadata_column[,1]))
num_levels <- length(column_levels)
color_levels <- col.wheel(num_levels)
levels(column_factors) <- color_levels
my_data.color[,1]<-as.character(column_factors)
return(my_data.color)
}
######################
# SUB(8): Create optimal contrast color selection using a color wheel
# adapted from https://stat.ethz.ch/pipermail/r-help/2002-May/022037.html
######################
col.wheel <- function(num_col, my_cex=0.75) {
cols <- rainbow(num_col)
col_names <- vector(mode="list", length=num_col)
for (i in 1:num_col){
col_names[i] <- getColorTable(cols[i])
}
cols
}
######################
######################
######################
# SUB(10): Convert all colors into format "#rrggbb"
# adapted from https://stat.ethz.ch/pipermail/r-help/2002-May/022037.html
######################
getColorTable <- function(col) {
rgb <- col2rgb(col);
col <- rgb2col(rgb);
sort(unique(col))
}
######################
######################
######################
# SUB(9): The inverse function to col2rgb()
# adapted from https://stat.ethz.ch/pipermail/r-help/2002-May/022037.html
######################
rgb2col <- function(rgb) {
rgb <- as.integer(rgb)
class(rgb) <- "hexmode"
rgb <- as.character(rgb)
rgb <- matrix(rgb, nrow=3)
paste("#", apply(rgb, MARGIN=2, FUN=paste, collapse=""), sep="")
}
######################
######################
# Create colors from the metadata
# generate metadata colors here
metadata_column = my_tidy_data[,'env_package.data.body_site', drop=FALSE]
metadata_column <- metadata_column[order(rownames(metadata_column)),,drop=FALSE]
# generate colors for selected column
metadata_colors <- create_colors(metadata_column)
# render 3d plot
plot_ly( x=my_tidy_data$V1,
y=my_tidy_data$V2,
z=my_tidy_data$V3,
type = "scatter3d",
mode = "markers",
#marker = list(color="red") # list(color = metadata_colors[,1]))
marker = list(color = metadata_colors[,1])
)