-
Notifications
You must be signed in to change notification settings - Fork 0
/
epiphytes_script.R
318 lines (249 loc) · 11.1 KB
/
epiphytes_script.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
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
rm(list=ls())
library(ggplot2)
library(iNEXT)
library(RColorBrewer)
library(ggpubr)
library(tableone)
library(tidyverse)
library(data.table)
library(rstatix)
library(vegan)
library(viridis)
#Data necessary to estimate the asymptotic diversity profile — only the empirical
df = read.csv('asymptotic_diversity_profile_estimated.csv',
sep = ';')
df = df %>% filter(Target == "Diversity")
df$Target <- gsub("Diversity", "Estimated diversity", df$Target)
#Plotting the asymptotic diversity profile
div_profile = ggplot(df, aes(x=Order.q, y=Estimate, colour=Community)) +
scale_x_continuous(breaks = c(0, 0.5, 1, 1.5, 2)) +
#scale_y_continuous(breaks = seq(0, 250, 25)) +
theme_bw() +
geom_point(aes(shape=Community), size = 1, data=df) +
geom_line(aes(linetype=Target), size = 1.5, data = df) +
geom_ribbon(aes(ymin=LCL, ymax=UCL,
fill=Community, colour=NULL), alpha=0.2) +
labs(x="Order of q", y="Species diversity") +
theme(legend.position = "bottom",
legend.title=element_blank(),
text=element_text(size=16),
legend.box = "vertical",
plot.title = element_text(size = 18)) +
guides(fill=guide_legend(nrow = 2, byrow = T)) +
scale_fill_manual(values = c("#B34436", "#FF8E80", '#24B36C', '#66FFB4')) +
scale_color_manual(values = c("#B34436", "#FF8E80", '#24B36C', '#66FFB4')) +
ggtitle("(a) Asymptotic diversity profile")
{
svg("diversity_profile_new.svg", width = 6, height= 8)
plot(div_profile)
dev.off()
}
#Abundance vectors of accidentals only and genuine epiphytes only, by host species
acc_als=c(30,10,10,6,5,4,3,3,3,3,2,2,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)
sum(acc_als)
acc_cya=c(40,23,14,13,6,5,4,4,4,4,4,3,3,3,2,2,2,2,2,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)
sum(acc_cya)
true_als=c(253,77,38,20,9,4,3,3,1)
sum(true_als)
true_cya=c(51,51,31,28,18,3,2,1,1,1)
sum(true_cya)
mylist <- list('True epiphytes on A. setosa' = true_als,
'True epiphytes on C. phalerata' = true_cya,
'Acc. epiphytes on A. setosa' = acc_als,
'Acc. epiphytes on C. phalerata' = acc_cya)
#Running Hill Numbers with Cmax (2 * minimum abundance vector or max. reference sample size -
#whichever is larger, Chao et al. 2014) — using the max reference sample of 408
res_mylist <-iNEXT(mylist, q = c(0,1,2), datatype = "abundance", nboot = 200, endpoint = 408)
##### RAREFACTION AND EXTRAPOLATION #####
#Fortifying dataframe for ggplot2 workaround
df <- ggplot2::fortify(res_mylist, type = 1)
head(df)
df.point <- df[which(df$method=="observed"),]
df.line <- df[which(df$method!="observed"),]
df.line$method <- factor(df.line$method,
c("interpolated", "extrapolated"),
c("interpolation", "extrapolation"))
div_plot <- ggplot(df, aes(x=x, y=y, colour=site)) +
theme_bw() +
geom_point(aes(shape=site), size=1, data=df.point) +
geom_line(aes(linetype=method), lwd=1.5, data=df.line) +
geom_ribbon(aes(ymin=y.lwr, ymax=y.upr,
fill=site, colour=NULL), alpha=0.2) +
facet_wrap(~order) +
labs(x="Number of individuals", y="Species diversity") +
theme(legend.position = "bottom",
legend.title=element_blank(),
text=element_text(size=16),
legend.box = "vertical",
plot.title = element_text(size = 18)) +
guides(fill=guide_legend(nrow = 2, byrow = T)) +
scale_fill_manual(values = c("#B34436", "#FF8E80", '#24B36C', '#66FFB4')) +
scale_color_manual(values = c("#B34436", "#FF8E80", '#24B36C', '#66FFB4')) +
ggtitle("(b) Non-asymptotic size-based rarefaction and extrapolation")
{
svg("rarext_plot.svg", width = 6, height= 8)
plot(div_plot)
dev.off()
}
##### MERGING FIGURES #####
figure_div <- ggpubr::ggarrange(div_profile, div_plot,
nrow = 2, ncol = 1, common.legend = F, legend = "bottom")
figure_div
{
png("new_plots.png", width = 8, height= 10, units = "in", res = 300)
par(mar=c(5, 1, 5, 1))
plot(figure_div)
dev.off()
}
##### QUASI-POISSON REGRESSION #####
#Loading and adjusting data (both genuine and accidental epiphytes)
acc <- read.csv('Biotropica_review/new_results/correlation_data_acc.csv',
header = TRUE, sep = ';')
names(acc) <- c("species", "humidity", "height", "cbh", "abundance", "richness")
#Obtaining radius
acc$radius = acc$cbh/pi
acc$radius = round(acc$radius, 2)
#Approximating to the area of a cilinder as A=2πrh+2πr^2
acc$area = (2*pi*acc$radius*acc$height)+(2*pi*(acc$radius^2))
#Are variables correlated?
cor(acc$int, acc$area)
#Interaction term to test later
acc$int <- acc$height*acc$humidity
#Standardizing variables
acc <- acc %>% mutate_at(c("humidity", "area", "int"), ~(scale(.) %>% as.vector))
#Since response variables are not normally distributed and data is overdispersed,
#we use the quasi-Poisson family distribution to fit the data
#performance::check_overdispersion(acc_model)
##### ABUNDANCE MODEL #####
ab_model <- glm(formula = abundance ~ height,
family = quasipoisson(link="log"), data = acc)
ab_model2 <- glm(formula = abundance ~ area + humidity + area*humidity,
family = quasipoisson(link="log"), data = acc)
#Significant difference between models?
anova(ab_model, ab_model2, test = "Chisq")
#Regression parameters for abundance model
tableone::ShowRegTable(ab_model)
#Pseudo R-squared for quasi-Poisson model: 1 - (residual.deviance/null.deviance)
1 - (ab_model$deviance/ab_model$null.deviance) #0.08 partial-R²
#Visualizing...
#Test for all variables separately
#New range area to predict
range(acc$int)
xrange <- seq(-1.4, 4, 0.25)
#Predicted model
yab <- predict(ab_model, list(int = xrange), type="response")
plot(acc$int, acc$abundance, pch = 16, xlab = "interaction", ylab = "ab")
lines(xrange, yab)
##### RICHNESS MODEL ######
rich_model <- glm(formula = richness ~ area + humidity + area*humidity,
family = quasipoisson(link="log"), data = acc)
#Regression parameters for richness model
tableone::ShowRegTable(rich_model)
#Pseudo R-squared for quasi-Poisson model: 1 - (residual.deviance/null.deviance)
1 - (rich_model$deviance/rich_model$null.deviance) #0.08 partial-R²
#Visualizing...
#Test for all variables separately
#New range area to predict
range(acc$int)
xrange <- seq(-1.4, 4, 0.25)
#Predicted model
yab <- predict(ab_model, list(int = xrange), type="response")
plot(acc$int, acc$abundance, pch = 16, xlab = "interaction", ylab = "ab")
lines(xrange, yab)
#### NMDS ANALYSIS #####
data <- read.csv('nmds_matrix_onlyacc.csv', header = TRUE, sep = ';')
spp <- data[, 2:ncol(data)]
group <- data$site
#Binarizing abundance data (Jaccard similarity index)
spp <- ifelse(spp > 0,1,0)
spp <- as.data.frame(spp)
#PERMANOVA to test if accidental composition differ between each host
adonis(formula = spp ~ group, permutations = 999, method = "jaccard")
#Running NMDS
nmds_results <- metaMDS(spp, k = 2, maxit = 999, trymax = 500,
wscore = TRUE, distance = "jaccard")
#Site data from NMDS object
data_scores <- as.data.frame(scores(nmds_results))
# Now add the extra aquaticSiteType column
data_scores <- cbind(data_scores, group)
colnames(data_scores)[3] <- "Host"
#Creating the hull object
grp.a <- data_scores[data_scores$Host == "A. setosa", ][chull(data_scores[data_scores$Host ==
"A. setosa", c("NMDS1", "NMDS2")]), ]
grp.b <- data_scores[data_scores$Host == "C. phalerata", ][chull(data_scores[data_scores$Host ==
"C. phalerata", c("NMDS1", "NMDS2")]), ]
hull.data <- rbind(grp.a, grp.b) #combine both hulls
hull.data
#Plotting
nmds_plot <- ggplot() +
geom_point(data = data_scores, aes(x = NMDS1, y = NMDS2,
color = Host), size = 2) +
geom_polygon(data = hull.data, aes(x = NMDS1, y = NMDS2, group = Host), alpha = 0.15) +
scale_color_manual(values = inferno(15)[c(3, 8)],
name = "Host") +
coord_equal() +
annotate(geom = "label", x = 0, y = -1.5, size = 8,
label = paste("Stress: ", round(nmds_results$stress, digits = 3))) +
theme_bw() +
theme(axis.text = element_text(size = 20),
legend.position = "bottom",
text = element_text(size = 18))
png("Biotropica_review/new_results/nmds.svg", width = 10, height= 8)
plot(nmds_plot)
dev.off()
##### STRATIFICATION GRAPH #####
data = read.csv('new_results/strat_graph_biotropica.csv', header = TRUE, sep = ';')
#data$class = factor(data$class)
setDT(data)
data[int%in% "A", int := "A (0-1m)"]
data[int%in% "B", int := "B (1-2m)"]
data[int%in% "C", int := "C (2-3m)"]
colorset = c("Non-Acc" = "yellowgreen", "Acc" = "darkorange")
abun_plot <- ggplot(data) +
geom_bar(aes(x = class, y = abun, fill = class), stat = "identity") + facet_wrap(~int) + theme_bw() +
scale_fill_manual(values=colorset) + guides(fill=F) + xlab("\nEcological category") +ylab("Abundance\n") +
theme(axis.text = element_text(size = 18), text = element_text(size = 20), axis.title.x = element_blank(),
axis.text.x = element_blank(),
axis.ticks.x = element_blank())
#Converting abundance to richness
data$richness = data[, c("abun")]
data$richness[data$richness > 0] <- 1
rich_plot <- ggplot(data) +
geom_bar(aes(x = class, y = richness, fill = class), stat = "identity") + facet_wrap(~int) + theme_bw() +
scale_fill_manual(values=colorset) + guides(fill=F) + xlab("\nEcological category") +ylab("Richness\n") +
theme(axis.text = element_text(size = 18), text = element_text(size = 20), strip.background = element_blank(),
strip.text.x = element_blank())
final_plot = ggarrange(abun_plot, rich_plot, ncol = 1, nrow = 2)
final_plot
ggsave('new_results/strat_graph.jpg', width = 10, height = 10)
dev.off()
##### HEIGHT-INTERVAL ANALYSIS #####
data = read.csv('strat_graph_biotropica.csv', header = TRUE, sep = ';')
#data$class = factor(data$class)
setDT(data)
data[int%in% "A", int := "Lower (0-1m)"]
data[int%in% "B", int := "Medium (1-2m)"]
data[int%in% "C", int := "Upper (2-3m)"]
data = subset(data, class == "Acc")
data = subset(data, abun > 0)
# Build the linear model
model <- lm(abun ~ int, data = data)
# Create a QQ plot of residuals
ggqqplot(residuals(model))
res.kruskal <- data %>% kruskal_test(abun ~ int)
res.kruskal
data %>% kruskal_effsize(abun ~ int)
# Pairwise comparisons
pwc <- data %>%
dunn_test(abun ~ int, p.adjust.method = "bonferroni")
pwc
# Visualization: box plots with p-values
pwc <- pwc %>% add_xy_position(x = "int")
ggboxplot(data, x = "int", y = "abun") +
stat_pvalue_manual(pwc, hide.ns = TRUE) +
labs(
subtitle = get_test_label(res.kruskal, detailed = TRUE),
caption = get_pwc_label(pwc)
) + ylab("Abundance") + xlab("Height class") + theme_bw()
ggsave("new_results/kruskal_newnames.png", height = 10, width = 10)
rm(list=ls())