-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathscript09_alpha-diver.R
291 lines (233 loc) · 9.88 KB
/
script09_alpha-diver.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
# ANALYZING ALPHA DIVERSITY #
# ------------------------- #
#--- Load objects needed for these steps ----
pqs <- readRDS(file = "./rds-files/pqs.rds")
#--- Define colors for graphics ----
colors.hab <- rev(brewer.pal(n = 11, name = "RdBu"))[c(3,9)]
### Calculating sequence abundance per sample ----
#--- Calculate abundances:
n.seqs.sample <- data.frame(cbind(sample_sums(pqs), get_variable(pqs, "Sponge_Species")))
colnames(n.seqs.sample)[1:2] <- c("Abundance", "Sample")
n.seqs.sample$Abundance <- as.numeric(n.seqs.sample$Abundance)
#--- Write table of total abundances:
write.table(n.seqs.sample, file = "./results/total-abundance.tsv",
sep = "\t", na = "NA", row.names = T, col.names = FALSE)
### Estimating the microbial diversity ----
#--- Calculate indexes:
adiv <- estimate_richness(pqs, split = TRUE,
measures = c("Observed", "Simpson", "Shannon"))
#--- Total ASVs (all samples):
sum(adiv[,'Observed'])
#--- Write table:
write.table(adiv, file = "./results/alpha-diversity.tsv",
sep = "\t", dec = ".",na = "NA", row.names = T, col.names = NA)
#--- Calculate Faith phylogenetic diversity:
faith <- pd(otu_table(pqs), phy_tree(pqs), include.root = T)
#--- Join alpha diversity data to metadata:
adiv.mdata <- data.table(metadata, faith, adiv, keep.rownames = FALSE)
### Performing LMM and GLMM ----
#--- For observed richness:
# Using the factor "Environment":
a <- glmer(SR ~ Environment +
# (1|Sponge_Order) +
(1|Sponge_Family) +
(1|Sequencing_run_ID), data = adiv.mdata, family = "poisson")
tab_model(a)
summary(a)
plot_model(a, type = "eff", terms = "Environment", show.data = T)
# Using the factor "Habitat":
b <- glmer(SR ~ Habitat +
# (1|Sponge_Order) +
(1|Sponge_Family) +
(1|Sequencing_run_ID), data = adiv.mdata, family = "poisson")
tab_model(b)
summary(b)
plot_model(b, type = "eff", terms = "Habitat", show.data = F)
#--- For Shannon index:
# Using the factor "Environment":
c <- lmer(Shannon ~ Environment +
# (1|Sponge_Order) +
(1|Sponge_Family) +
(1|Sequencing_run_ID), data = adiv.mdata)
tab_model(c)
summary(c)
plot_model(c, type = "eff", terms = "Environment", show.data = T)
# Using the factor "Habitat":
d <- lmer(Shannon ~ Habitat +
# (1|Sponge_Order) +
(1|Sponge_Family) +
(1|Sequencing_run_ID), data = adiv.mdata)
tab_model(d)
summary(d)
plot_model(d, type = "eff", terms = "Habitat", show.data = F)
adiv.mdata$jost <- exp(adiv.mdata$Shannon) # Alternative
d1 <- lmer(jost ~ Habitat +
(1|Sponge_Family) +
(1|Sequencing_run_ID), data = adiv.mdata)
tab_model(d1)
plot_model(d1, type = "eff", terms = "Habitat", show.data = F)
#--- For Simpson index:
# Using the factor "Environment":
e <- lmer(Simpson ~ Environment +
# (1|Sponge_Order) +
(1|Sponge_Family) +
(1|Sequencing_run_ID), data = adiv.mdata)
tab_model(e)
plot_model(e, type = "eff", terms = "Environment", show.data = T)
f <- lmer(Simpson ~ Habitat +
# (1|Sponge_Order) +
(1|Sponge_Family) +
(1|Sequencing_run_ID), data = adiv.mdata)
tab_model(f)
summary(f)
plot_model(f, type = "eff", terms = "Habitat", show.data = T)
#--- For Faith PD:
pd <- lmer(PD ~ Habitat +
(1|Sponge_Family) +
(1|Sequencing_run_ID), data = adiv.mdata)
tab_model(pd)
plot_model(pd, type = "eff", terms = "Habitat", show.data = F)
### Comparing models ----
#--- Compare models with Akaike:
aictab(cand.set = list(a, b), modnames = c("a", "b"))
aictab(cand.set = list(c, d), modnames = c("c", "d"))
aictab(cand.set = list(e, f), modnames = c("e", "f"))
### Plotting microbial diversity ----
#--- Bring results from LMM and GLMM:
rich.result <- ggpredict(model = b, terms = "Habitat")
shan.result <- ggpredict(model = d, terms = "Habitat")
simp.result <- ggpredict(model = f, terms = "Habitat")
fait.result <- ggpredict(model = pd, terms = "Habitat")
#--- Previous plots:
plot(rich.result)
plot(shan.result)
plot(simp.result)
plot(fait.result)
#--- Customized bar plots:
# For observed richness:
rich.bar <- ggplot(rich.result, aes(x = x, y = predicted, fill = x)) +
geom_bar(stat = "identity", color = "black",
position = position_dodge()) +
geom_errorbar(aes(x = x, ymin = conf.low, ymax = conf.high, y = predicted),
width = 0.4) +
geom_expected_point(aes(x = x, y = predicted), show.legend = FALSE) +
theme_classic() +
labs(x = NULL, y = "Observed richness", fill = NULL) +
scale_fill_manual(values = colors.hab) +
theme(legend.position = "none", axis.text.x = element_blank(),
axis.ticks.x = element_blank(), axis.text.y = element_text(size = 15),
text = element_text(size = 18))
# For Shannon index:
shan.bar <- ggplot(shan.result, aes(x = x, y = predicted, fill = x)) +
geom_bar(stat = "identity", color = "black",
position = position_dodge()) +
geom_errorbar(aes(x = x, ymin = conf.low, ymax = conf.high, y = predicted),
width = 0.4) +
geom_expected_point(aes(x = x, y = predicted), show.legend = FALSE) +
theme_classic() +
labs(x = NULL, y = "Shannon", fill = NULL) +
scale_fill_manual(values = colors.hab) +
theme(legend.position = "bottom", axis.text.x = element_blank(),
axis.ticks.x = element_blank(), legend.text = element_text(size = 15),
axis.text.y = element_text(size = 15),
text = element_text(size = 18))
# For Simpson index:
simp.bar <- ggplot(simp.result, aes(x = x, y = predicted, fill = x)) +
geom_bar(stat = "identity", color = "black",
position = position_dodge()) +
geom_errorbar(aes(x = x, ymin = conf.low, ymax = conf.high, y = predicted),
width=0.4) +
geom_expected_point(aes(x = x, y = predicted), show.legend = FALSE) +
theme_classic() +
labs(x = NULL, y = "Simpson", fill = NULL) +
scale_fill_manual(values = colors.hab)+
theme(legend.position = "none", axis.text.x = element_blank(),
axis.ticks.x = element_blank(), axis.text.y = element_text(size = 15),
text = element_text(size = 18))
fait.bar <- ggplot(fait.result, aes(x = x, y = predicted, fill = x)) +
geom_bar(stat = "identity", color = "black",
position = position_dodge()) +
geom_errorbar(aes(x = x, ymin = conf.low, ymax = conf.high, y = predicted),
width=0.4) +
geom_expected_point(aes(x = x, y = predicted), show.legend = FALSE) +
theme_classic() +
labs(x = NULL, y = "Faith", fill = NULL) +
scale_fill_manual(values = colors.hab)+
theme(legend.position = "none", axis.text.x = element_blank(),
axis.ticks.x = element_blank(), axis.text.y = element_text(size = 15),
text = element_text(size = 18))
#--- Save HQ image (also can use ggsave() or png() ):
png("./alpha-rich.png", units = "in", width = 3, height = 3, res = 300)
rich.bar
dev.off()
png("./alpha-simp.png", units = "in", width = 3, height = 3, res = 300)
simp.bar
dev.off()
png("./alpha-shan.png", units = "in", width = 3, height = 3, res = 300)
shan.bar
dev.off()
#--- Plot just the legend:
legend <- cowplot::get_legend(shan.bar)
legend <- as_ggplot(legend)
png("./alpha-legend.png", units = "in", width = 9, height = 0.5, res = 300)
legend #width = 2, height = 3
dev.off()
#--- Scatter plots with error bars:
# rich.plot <- ggplot() +
# geom_point(data = adiv.mdata,
# aes(x = Habitat, y = SR, color = Habitat),
# position = position_dodge2(0.7)) +
# geom_errorbar(data = rich.result,
# aes(x = x, ymin = conf.low, ymax = conf.high, y = predicted)) +
# geom_expected_point(data = c.result,
# aes(x = x, y = predicted)) +
# labs(x = NULL, y = "Species richness", color = NULL) +
# theme_classic() +
# theme(legend.position = "right",
# axis.text.x = element_blank(), axis.ticks.x = element_blank()) +
# scale_color_manual(values = colors.hab)
#
# shan.plot <- ggplot() +
# geom_point(data = adiv.mdata,
# aes(x = Habitat, y = Shannon, color = Habitat),
# position = position_dodge2(0.7)) +
# geom_errorbar(data = shan.result,
# aes(x = x, ymin = conf.low, ymax = conf.high, y = predicted)) +
# geom_expected_point(data = d.result,
# aes(x = x, y = predicted)) +
# labs(x = NULL, y = "Shannon", color = NULL) +
# theme_classic() +
# theme(legend.position = "right",
# axis.text.x = element_blank(), axis.ticks.x = element_blank()) +
# scale_color_manual(values = colors.hab)
#
# simp.plot <- ggplot() +
# geom_point(data = adiv.mdata,
# aes(x = Habitat, y = Simpson, color = Habitat),
# position = position_dodge2(0.7)) +
# geom_errorbar(data = simp.result,
# aes(x = x, ymin = conf.low, ymax = conf.high, y = predicted)) +
# geom_expected_point(data = k.result,
# aes(x = x, y = predicted)) +
# labs(x = NULL, y = "Simpson", color = NULL) +
# theme_classic() +
# theme(legend.position = "right",
# axis.text.x = element_blank(), axis.ticks.x = element_blank()) +
# scale_color_manual(values = colors.hab)
### Pattern of biogeography over richness (bonus) ----
#--- Plot richness across latitudes:
rich.biog <- ggplot(data = adiv.mdata, aes(x = Observed, y = Latitude)) +
geom_point() +
scale_y_continuous(breaks = seq(-80, 80, 20)) +
geom_smooth(col = "red", size = 1, orientation = "y")
#--- Save HQ image (also can use ggsave() or png() ):
png("./results/graphics/rich-biogeo.png", units = "in", width = 7, height = 5, res = 300)
rich.biog
dev.off()
### Save things ------------------------------------------------------------
#--- Save objects:
saveRDS(adiv.mdata, file = "./rds-files/adiv.mdata.rds")
saveRDS(n.seqs.sample, file = "./rds-files/n_seqs_sample.rds")
saveRDS(adiv, file = "./rds-files/alpha_diversity.rds")
#---- Save the current work and objects:
save.image("R-graph-adiv.RData")