forked from schraderL/GAGA
-
Notifications
You must be signed in to change notification settings - Fork 0
/
ancestral_reconstruction_script.R
242 lines (197 loc) · 12 KB
/
ancestral_reconstruction_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
library(treeio)
library(ggtree)
library(phytools)
library(ggplot2)
library(ggnewscale)
library(wesanderson)
library(ggimage)
library(geiger)
setwd("/Users/joel/GAGAproject/Ancestral_reconstruction/")
### READ TREE
treeu <- read.tree("GAGA_dated_phylogeny_newick.tre")
## READ TRAITS TABLE
traits <- read.csv("Traits_table_v6_18Dec_toPGLS_reduced.tsv",sep="\t",dec = ".", quote = "")
# READ SUBFAMILIES & NODES TABLE: to plot the subfamily ranges with the corresponding colors from ITOL
subfamilies <- read.csv("subfamilies_nodes.csv",sep="\t",dec = ",")
subfamilies_colors <- c("#04aa08", "#c80a0a","#020bf3","#e77a01","#1cc2c4","#60c304","#03783f","#d2ae04","#4505d4","#6d3c12","#f41e23","#069ffe")
names(subfamilies_colors) <- subfamilies$Subfamily
### Phylogenetic signal
traittotest <- traits$Colony.size.cat
names(traittotest) <- traits$GAGA.ID
phylosig(treeu, traittotest, method = "lambda", test = TRUE)
lambda.traittotest<-fitContinuous(phy = treeu, dat = traittotest, model = "lambda")
lambda.traittotest
#################### DISCRETE CHARACTERS: Ancestral Character Estimation ##################################################
#The model of trait evolution used for each trait is determined by comparing the log likelihoods of alternative models
TRAIT <- traits$Colony.size.cat # Select here the column with trait of interest to conduct the ancestral reconstruction
names(TRAIT) <- paste(traits$GAGA.ID, traits$Species.name) # Retrieve GAGA ID for each trait and the species name
TRAIT<- replace(TRAIT, TRAIT=="", NA)
unique(TRAIT) # See the categories for the trait
numtraits <- 4 # Write here the number of traits (do not count NA)
traitname <- "Colony size" # Name for the trait
TRAIT_colors <- as.vector(wes_palette("Zissou1", numtraits, type = "continuous")) # colors for the plot
# Visualize the color chosen
plot(NULL, xlim=c(0,length(TRAIT_colors)), ylim=c(0,1),
xlab="", ylab="", xaxt="n", yaxt="n")
rect(0:(length(TRAIT_colors)-1), 0, 1:length(TRAIT_colors), 1, col=TRAIT_colors)
names(TRAIT_colors) <- c("0", "1", "2", "3") # Name of the categories
# rename tree labels as "GAGA-ID species name"
for (name in 1:length(treeu$tip.label)) {
treeu$tip.label[name] <- paste(treeu$tip.label[name],
traits$Species.name[which(traits$GAGA.ID == treeu$tip.label[name])])
}
# Ancestral character estimation, testing models
fitER<-ace(TRAIT, treeu, model="ER", type="discrete") # Ancestral Character Estimation: ER model
fitARD<-ace(TRAIT, treeu, model="ARD", type="discrete") # Ancestral Character Estimation: ARD model
fitSYM<-ace(TRAIT, treeu, model="SYM", type="discrete") # Ancestral Character Estimation: SYM model
# Using other alternative models to test more relevant biological hypotheses for specific trait catgorizations
fit2single<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 0, 1, 0), 2)) # Ancestral Character Estimation
fit2singlerev<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 1, 0, 0), 2), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitpp<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 4, 0, 0,
4, 0, 5, 0,
0, 3, 0, 0,
0, 2, 1, 0), 4), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitpp<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 1, 2, 3, 4, 5, 0,
1, 0, 7, 8, 9, 10, 0,
2, 7, 0, 12, 13, 14, 0,
3, 8, 12, 0, 16, 17, 0,
4, 9, 13, 16, 0, 19, 0,
5, 10, 14, 17, 19, 0, 0,
6, 11, 15, 18, 20, 21, 0), 7), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitpp<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 1, 2,
1, 0, 3,
0, 0, 0), 3), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitpp2<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 1, 2,
0, 0, 3,
0, 0, 0), 3), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitmanc<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 0, 0,
1, 0, 3,
2, 3, 0), 3), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitmancard<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 4, 0,
1, 0, 5,
2, 3, 0), 3), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitmancard2<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 4, 0,
1, 0, 0,
2, 3, 0), 3), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitmancsym<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 1, 0,
1, 0, 0,
2, 3, 0), 3), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitmanc2<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 0, 0,
1, 0, 0,
2, 3, 0), 3), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
fitpp6<-ace(TRAIT, treeu, type="discrete", model=matrix(c(0, 0, 0, 0, 0, 0,
1, 0, 6, 7, 8, 9,
2, 6, 0, 10, 11, 12,
3, 7, 10, 0, 13, 14,
4, 8, 11, 13, 0, 15,
5, 9, 12, 14, 15, 0), 6), marginal = TRUE, use.expm = FALSE) # Ancestral Character Estimation
# Checking likelihoods and AIC's
#fitER$loglik
AIC(fitER)
#fitARD$loglik
AIC(fitARD)
#fitSYM$loglik
AIC(fitSYM)
#fit2single$loglik
AIC(fit2single)
#fit2ard$loglik
AIC(fit2ard)
#fitpp$loglik
AIC(fitpp)
AIC(fitmancsym)
AIC(fitmancard2)
# Type here the best retrieved model (lowest AIC)
best_model <- fitER # type here the best model
# Creating data frame with the reconstructed states for the best model
mas <- as.data.frame(round(best_model$lik.anc,2))
#mas <- as.data.frame(round(best_model$lik.anc, digits = 0))
mas$node <- row.names(mas)
# Plotting the tree
p1 <-ggtree(treeu, size = 0.2) +
xlim(0,300) +
geom_tiplab(offset=10,align= TRUE, size=0.9, linesize = 0.2)+
# geom_hilight(data=subfamilies, mapping = aes(node=node,fill=Subfamily, ), align = "right", type = "gradient", alpha=0.3)+
geom_hilight(data=subfamilies, mapping = aes(node=node,fill=Subfamily, ), align = "right", type = "rect", alpha=0.2)+
scale_fill_manual(values=subfamilies_colors) + guides(fill = guide_legend(override.aes = list(alpha = 1)))+
scale_y_reverse()+ new_scale_fill()
p1
# add heatmap to the tree (column at the right of the tree with the trait values)
p2 <- gheatmap(p1, data.frame(factor(TRAIT)), width=0.02, low="#D95F02",
high="#7570B3", colnames = FALSE, font.size=2, color="white", offset=4) +
scale_fill_manual(values=TRAIT_colors, labels = names(TRAIT_colors), name = traitname, na.value = NA)+
scale_y_reverse()
p2
# Add the pie charts with the ancestral reconstructions
pies <- nodepie(mas, cols=1:numtraits, color=TRAIT_colors, alpha=0.8)
#p12 <- inset(p2, pies, width = 0.06, height = 0.06, reverse_y = TRUE, hjust = 0.005, vjust = -0.2)
#p12 <- inset(p2, pies, width = 0.06, height = 0.06, reverse_y = TRUE, hjust = 1, vjust = -0.4)
p12 <- inset(p2, pies, width = 0.09, height = 0.09, reverse_y = TRUE, hjust = 1, vjust = -0.4) # Mac updated to R 4.2.2- change the scale
p12
# Save the tree into a file
ggsave(
"TraitNNN_ult_ancestral_fitER_discrete.pdf",
device = "pdf",
plot = last_plot(),
height = 6.33,
width = 5.14,
units = "in"
)
dev.off()
#################### ANCESTRAL RECONSTRUCTION FOR CONTINUOUS CHARACTERS ####################################################################
### Note: the continuous trait ancestral reconstruction does not accepted missing values or NA
### Then, if the dataset contains missing data, those species need to be excluded from the table and tree, here the commands to do it:
## Subset tree to only contain species without NA (here as example the trait column MATmean)
#tree.subset<-treeio::drop.tip(treeu@phylo,tip=traits$GAGA.ID[is.na(traits$MATmean)])
## trim trait data to same set of species
#TRAIT<-traits$MATmean[!is.na(traits$MATmean)]
#names(TRAIT)<-traits$GAGA.ID[!is.na(traits$MATmean)]
setwd("/Users/joel/GAGAproject/Ancestral_reconstruction/")
treeu <- read.tree("GAGA_dated_phylogeny_newick.tre")
traits <- read.csv("Traits_table_v6_18Dec_toPGLS_reduced.tsv",sep="\t",dec = ".", quote = "")
subfamilies <- read.csv("subfamilies_nodes.csv",sep="\t",dec = ",")
subfamilies_colors <- c("#04aa08", "#c80a0a","#020bf3","#e77a01","#1cc2c4","#60c304","#03783f","#d2ae04","#4505d4","#6d3c12","#f41e23","#069ffe")
names(subfamilies_colors) <- subfamilies$Subfamily
TRAIT_colors <- as.vector(wes_palette("Zissou1", 5, type = "continuous")) # change by number of categories/colours to plot the trait
TRAIT <- traits$Colony.size.log.max # select column with CONTINUOUS trait of interest
#TRAIT <- as.numeric(gsub(",","",traits$Genome.size)) # select column with CONTINUOUS trait of interest
traitname <- "Colony"
names(TRAIT) <-traits$GAGA.ID # obtain GAGA ID for each trait
## Ancestral reconstruction for the continuous trait
fit<-fastAnc(treeu, TRAIT) # fast estimation of ML ancestral states
# Parsing the results
traitvector <- data.frame(node = ggtree::nodeid(treeu, names(TRAIT)), trait = TRAIT)
traitfit <- data.frame(node = names(fit), trait = as.vector(fit))
traitfit2 <- rbind(traitvector, traitfit)
traitfit2$node<-as.numeric(traitfit2$node)
tree2 <- as.treedata(tidytree::full_join(as_tibble(treeu), traitfit2, by = 'node'))
# rename tree labels in format "GAGAID species name"
for (name in 1:length(tree2@phylo$tip.label)) {
tree2@phylo$tip.label[name] <- paste(tree2@phylo$tip.label[name],
traits$Species.name[which(traits$GAGA.ID == tree2@phylo$tip.label[name])])
}
# Plot the tree
p1 <- ggtree(tree2,ladderize = T, aes(color=trait), continuous = "color", size=0.3,
lineend='square')+ scale_y_reverse() +
geom_tiplab(offset=1,align = TRUE ,color = "black", size = 1, linesize = 0.2)+
scale_color_gradientn(colours=TRAIT_colors, name = traitname)+
# scale_color_gradientn(colours=TRAIT_colors, labels = names(TRAIT_colors), name = "Colony size")+
coord_cartesian(clip = 'off') +
xlim(0, 300)+ theme(legend.position = "left")
# Add subfamily to the right of the tree
#for (i in 1:length(subfamilies_colors)) { # Changed offset here to use in my home mac
# p1 <- p1 + geom_cladelabel(align = TRUE,subfamilies$node[i], names(subfamilies_colors[i]), offset=80, barsize=0.5,
# angle=0, offset.text=1, hjust=0, fontsize=1, color = subfamilies_colors[i])
#}
for (i in 1:length(subfamilies_colors)) {
p1 <- p1 + geom_cladelabel(align = TRUE,subfamilies$node[i], names(subfamilies_colors[i]), offset=100, barsize=0.5,
angle=0, offset.text=1, hjust=0, fontsize=1, color = subfamilies_colors[i])
}
p1
ggsave(
"Colony_size_log_max_ancestral_continuous.pdf",
device = "pdf",
plot = last_plot(),
height = 6.33,
width = 5.14,
units = "in"
)