-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathFigure7.R
240 lines (168 loc) · 11.7 KB
/
Figure7.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
rm(list=ls());gc()
col_GA <- c("cadetblue1", "cornflowerblue", "slateblue2")
load("output_files/RData/G_matrices_high_salt.RData")
sampled_variance_NaCl=NULL
sampled_variance_GA1=NULL
sampled_variance_GA2=NULL
sampled_variance_GA4=NULL
for(i in 1:nrow(VCV_mat_NaCl[[1]]$VCV_Mat)){
sampled_variance_NaCl=rbind(sampled_variance_NaCl,c(sum(eigen(matrix(VCV_mat_NaCl[[1]]$VCV_Mat[i,1:49],7,7)/2)$values),eigen(matrix(VCV_mat_NaCl[[1]]$VCV_Mat[i,1:49],7,7)/2)$values))
sampled_variance_GA1 =rbind(sampled_variance_GA1,c(sum(eigen(matrix(VCV_mat_NaCl[[2]]$VCV_Mat[i,1:49],7,7)/2)$values),eigen(matrix(VCV_mat_NaCl[[2]]$VCV_Mat[i,1:49],7,7)/2)$values))
sampled_variance_GA2 =rbind(sampled_variance_GA2,c(sum(eigen(matrix(VCV_mat_NaCl[[3]]$VCV_Mat[i,1:49],7,7)/2)$values),eigen(matrix(VCV_mat_NaCl[[3]]$VCV_Mat[i,1:49],7,7)/2)$values))
sampled_variance_GA4 =rbind(sampled_variance_GA4,c(sum(eigen(matrix(VCV_mat_NaCl[[4]]$VCV_Mat[i,1:49],7,7)/2)$values),eigen(matrix(VCV_mat_NaCl[[4]]$VCV_Mat[i,1:49],7,7)/2)$values))
}
all_sampled_variance=list(sampled_variance_NaCl,sampled_variance_GA1,sampled_variance_GA2,sampled_variance_GA4)
load('output_files/RData/Random_skewers.RData')
pdf(file='plots/Figure7.pdf',h=8,w=7.5)
layout(matrix(c(1,1,2,3),2,2),w=c(1.2,1))
par(mar=c(5,7,4,2))
vect_Var <- c(2:7,10:14,18:21,26:28,34,35,42,1,9,17,25,33,41,49)
vProb <- .95
plot(c(VCV_mat_NaCl[[1]]$G1_mat/2)[vect_Var],c(31:11,7:1),yaxt="n",bty="n",xlim=c(-.25,.50),xlab="Genetic (co)variances",xaxt="n",type='n',ylab="",cex.lab=1.2)
mtext(side=3,at=-.5,"A",cex=1.5)
lines(c(0,0),c(31.5,8.5))
lines(c(0,0),c(.5,7.5))
axis(side=1,pos=0)
axis(side=2,at=c(31:11,7:1),labels=c("SF*SB","SF*FS","SF*FB","SF*BS","SF*BF","SF*Size",
"SB*FS","SB*FB","SB*BS","SB*BF","SB*Size",
"FS*FB","FS*BS","FS*BF","FS*Size",
"FB*BS","FB*BF","FB*Size","BS*BF","BS*Size","BF*Size",
"SF","SB","FS","FB","BS","BF","Size"),las=1)
for(i in 1:4){
temp_95 <- HPDinterval(VCV_mat_NaCl[[i]]$VCV_Mat[,1:49]/2,prob=.95)
arrows(temp_95[vect_Var,1],c(31:11,7:1)+(.15*(i-1)),temp_95[vect_Var,2],c(31:11,7:1)+(.15*(i-1)),code=3,length=.02,angle=90)
temp_80 <- HPDinterval(VCV_mat_NaCl[[i]]$VCV_Mat[,1:49]/2,prob=.83)
arrows(temp_80[vect_Var,1],c(31:11,7:1)+(.15*(i-1)),
temp_80[vect_Var,2],c(31:11,7:1)+(.15*(i-1)),code=3,length=0,angle=90,lwd=2,col=c("grey",col_GA)[i])
points(c(VCV_mat_NaCl[[i]]$G1_mat/2)[vect_Var],c(31:11,7:1)+(.15*(i-1)),pch=21,bg="black",cex=.6)
}
legend(.15,23,c("A6140","GA150","GA250","GA450"),lwd=2,bty="s",col=c("grey",col_GA))
par(mar=c(5,4,4,2))
plot(c(0,1),c(
sum(eigen(VCV_mat_NaCl[[1]]$G1_mat/2)$values),
sum(eigen(VCV_mat_NaCl[[1]]$G1_mat/2)$values)),bg=c("black","black"),pch=21,ylab='Total genetic variance',xlim=c(0,1.3),ylim=c(0,.83),type="n",bty="n",xaxt="n",xlab="")
#axis(side=1,at=c(0,1),labels=c("A6140","GA[1,2,4]50"),padj=.5)
mtext(side=3,at=-.5,"B",cex=1.5)
posX=c(0,.7,1,1.3)
axis(side=1,at=posX,c("A6140","GA150","GA250","GA450"),las=2)
for(i in 1:4){
temp_int <- HPDinterval(mcmc(all_sampled_variance[[i]][,1]),prob=.95)
arrows(posX[i],temp_int[1],posX[i],temp_int[2],length=.05,code=3,angle=90)
temp_int <- HPDinterval(mcmc(all_sampled_variance[[i]][,1]),prob=.83)
arrows(posX[i],temp_int[1],posX[i],temp_int[2],length=0,code=3,angle=90,lwd=2,col=c("grey",col_GA)[i])
}
points(posX,c(
sum(eigen(VCV_mat_NaCl[[1]]$G1_mat/2)$values),
sum(eigen(VCV_mat_NaCl[[2]]$G1_mat/2)$values),sum(eigen(VCV_mat_NaCl[[3]]$G1_mat/2)$values),sum(eigen(VCV_mat_NaCl[[4]]$G1_mat/2)$values)),bg=c("grey",col_GA),pch=21)
# Variance along e_max
plot(posX,c(0,0,0,2),type="n",las=1,bty="n",ylab=expression(paste("Genetic variance along ",e[max])),xaxt="n",xlab="")
mtext(side=3,at=-.5,"C",cex=1.5)
axis(side=1,at=posX,c("A6140","GA150","GA250","GA450"),las=2)
arrows(posX,HPD.R.vec.proj[,1,1],posX,HPD.R.vec.proj[,2,1],code=3,angle=90,length=0.05)
arrows(posX,HPD.R.vec.proj.83[,1,1],posX,HPD.R.vec.proj.83[,2,1],length=0,col=c("grey",col_GA),lwd=2)
points(posX,R.postmode[1,],pch=16)
dev.off()
# Load the low salt G matrices
load("output_files/RData/G_matrices_low_salt.RData")
## Load the randomized G matrices
null_matrix=new.env()
load("output_files/RData_Random/Random_G_Analysis_Cemee_Pop_WI_A6140_NaCl_LIGHT.RData",envir=null_matrix)
A6140_NaCl_null = null_matrix$df_G1
load("output_files/RData_Random/Random_G_Analysis_Cemee_Pop_WI_A6140_NGM_LIGHT.RData",envir=null_matrix)
A6140_NGM_null = null_matrix$df_G1
rm(null_matrix);gc()
sampled_variance_NGM_null=NULL
sampled_variance_NaCl_null=NULL
sampled_variance_NGM_null_v2=NULL
sampled_variance_NaCl_null_v2=NULL
Lambda_NGM <- eigen(VCV_mat_NGM[[1]]$G1_mat/2)$vectors
Lambda_NaCl <- eigen(VCV_mat_NaCl[[1]]$G1_mat/2)$vectors
for(i in 1:nrow(A6140_NGM_null)){
sampled_variance_NGM_null =rbind(sampled_variance_NGM_null,c(sum(eigen(matrix(A6140_NGM_null[i,],7,7)/2)$values),eigen(matrix(A6140_NGM_null[i,],7,7)/2)$values))
sampled_variance_NaCl_null=rbind(sampled_variance_NaCl_null,c(sum(eigen(matrix(A6140_NaCl_null[i,],7,7)/2)$values),eigen(matrix(A6140_NaCl_null[i,],7,7)/2)$values))
## Rotate the matrix along the true eigenvectors
sampled_variance_NGM_null_v2 = rbind(sampled_variance_NGM_null_v2, diag(t(Lambda_NGM) %*% (matrix(A6140_NGM_null[i,],7,7)/2) %*% Lambda_NGM))
sampled_variance_NaCl_null_v2 = rbind(sampled_variance_NaCl_null_v2, diag(t(Lambda_NaCl) %*% (matrix(A6140_NaCl_null[i,],7,7)/2) %*% Lambda_NaCl))
}
### Figure Supplement
### We need the null G matrices of the GA[1-4] populations
null_matrix=new.env()
load("output_files/RData_Random/Random_G_Analysis_Cemee_Pop_WI_GA150_NaCl_LIGHT.RData",envir=null_matrix)
GA150_NaCl_null = null_matrix$df_G1
load("output_files/RData_Random/Random_G_Analysis_Cemee_Pop_WI_GA250_NaCl_LIGHT.RData",envir=null_matrix)
GA250_NaCl_null = null_matrix$df_G1
load("output_files/RData_Random/Random_G_Analysis_Cemee_Pop_WI_GA450_NaCl_LIGHT.RData",envir=null_matrix)
GA450_NaCl_null = null_matrix$df_G1
rm(null_matrix);gc()
sampled_variance_GA1_null=NULL
sampled_variance_GA2_null=NULL
sampled_variance_GA4_null=NULL
sampled_variance_GA1_null_v2=NULL
sampled_variance_GA2_null_v2=NULL
sampled_variance_GA4_null_v2=NULL
Lambda_GA1 <- eigen(VCV_mat_NaCl[[2]]$G1_mat/2)$vectors
Lambda_GA2 <- eigen(VCV_mat_NaCl[[3]]$G1_mat/2)$vectors
Lambda_GA4 <- eigen(VCV_mat_NaCl[[4]]$G1_mat/2)$vectors
for(i in 1:nrow(GA150_NaCl_null)){
sampled_variance_GA1_null =rbind(sampled_variance_GA1_null,c(sum(eigen(matrix(GA150_NaCl_null[i,],7,7)/2)$values),eigen(matrix(GA150_NaCl_null[i,],7,7)/2)$values))
sampled_variance_GA2_null =rbind(sampled_variance_GA2_null,c(sum(eigen(matrix(GA250_NaCl_null[i,],7,7)/2)$values),eigen(matrix(GA250_NaCl_null[i,],7,7)/2)$values))
sampled_variance_GA4_null =rbind(sampled_variance_GA4_null,c(sum(eigen(matrix(GA450_NaCl_null[i,],7,7)/2)$values),eigen(matrix(GA450_NaCl_null[i,],7,7)/2)$values))
## Rotate the matrix along the true eigenvectors of A6140
sampled_variance_GA1_null_v2 = rbind(sampled_variance_GA1_null_v2, diag(t(Lambda_NaCl) %*% (matrix(GA150_NaCl_null[i,],7,7)/2) %*% Lambda_NaCl))
sampled_variance_GA2_null_v2 = rbind(sampled_variance_GA2_null_v2, diag(t(Lambda_NaCl) %*% (matrix(GA250_NaCl_null[i,],7,7)/2) %*% Lambda_NaCl))
sampled_variance_GA4_null_v2 = rbind(sampled_variance_GA4_null_v2, diag(t(Lambda_NaCl) %*% (matrix(GA450_NaCl_null[i,],7,7)/2) %*% Lambda_NaCl))
}
true_GA1_v2 <- diag(t(Lambda_NaCl) %*% (VCV_mat_NaCl[[2]]$G1_mat/2) %*% Lambda_NaCl)
true_GA2_v2 <- diag(t(Lambda_NaCl) %*% (VCV_mat_NaCl[[3]]$G1_mat/2) %*% Lambda_NaCl)
true_GA4_v2 <- diag(t(Lambda_NaCl) %*% (VCV_mat_NaCl[[4]]$G1_mat/2) %*% Lambda_NaCl)
pdf(file='plots/Figure7_figure_supplement1.pdf')#,h=9,w=4)
temp_95_NULL_GA1 <- HPDinterval(as.mcmc(GA150_NaCl_null/2),prob=.95)
temp_95_NULL_GA2 <- HPDinterval(as.mcmc(GA250_NaCl_null/2),prob=.95)
temp_95_NULL_GA4 <- HPDinterval(as.mcmc(GA450_NaCl_null/2),prob=.95)
vect_CoVar <- c(2:7,10:14,18:21,26:28,34,35,42)
vect_Diag <- c(1,9,17,25,33,41,49)
plot(1:7,c(VCV_mat_NaCl[[2]]$G1_mat/2)[vect_Diag],pch=21,col="black",las=1,bty="n",ylab=c("Genetic Variance"),ylim=c(0,.2),type="n",xlim=c(0.7,7),xaxt="n",xlab="Transition rates")
axis(side=1,at=1:7,labels=c("SF","SB","FS","FB","BS","BF","Size"))
arrows(1:7-.15,temp_95_NULL_GA1[vect_Diag,1],1:7-.15,temp_95_NULL_GA1[vect_Diag,2],code=3,length=.05,col="orange",angle=90,lwd=2)
points(1:7-.15,c(VCV_mat_NaCl[[2]]$G1_mat/2)[vect_Diag],pch=21,col="black",bg=col_GA[1])
arrows(1:7,temp_95_NULL_GA2[vect_Diag,1],1:7,temp_95_NULL_GA2[vect_Diag,2],code=3,length=.05,col="orange",angle=90,lwd=2)
points(1:7,c(VCV_mat_NaCl[[3]]$G1_mat/2)[vect_Diag],pch=21,col="black",bg=col_GA[2])
arrows(1:7+.15,temp_95_NULL_GA4[vect_Diag,1],1:7+.15,temp_95_NULL_GA4[vect_Diag,2],code=3,length=.05,col="orange",angle=90,lwd=2)
points(1:7+.15,c(VCV_mat_NaCl[[4]]$G1_mat/2)[vect_Diag],pch=21,col="black",bg=col_GA[3])
legend(.9,.21,"Posterior mode",bty="n")
legend(1.15,.2,c("GA150","GA250","GA450"),bty="n",ncol=3,pch=21,pt.bg=col_GA,col="black")
legend(.9,.19,lwd=2,"95% CI of null posterior\nmodes",bty="n",col="orange")
dev.off()
pdf(file='plots/Figure7_figure_supplement2.pdf')#,h=9,w=4)
plot(eigen(VCV_mat_NaCl[[1]]$G1_mat/2)$values~c(1:7),pch=16,xlim=c(1,7.5),bty="n",ylim=c(0.001,.83),ylab=expression(paste('Genetic variance (',lambda[i],")")),type="n",xaxt="n",xlab="",log="y")
axis(1,at=1:7,c(expression(g[max]),expression(g[2]),expression(g[3]),expression(g[4]),expression(g[5]),expression(g[6]),expression(g[7])),las=1)
temp_int <- HPDinterval(mcmc(sampled_variance_GA1_null[,2:8]),prob=.95)
arrows((1:7)-.15,temp_int[,1],(1:7)-.15,temp_int[,2],length=.05,code=3,angle=90,col="orange",cex=1.2)
points(c(1:7)-.15,eigen(VCV_mat_NaCl[[2]]$G1_mat/2)$values ,pch=21,bg=col_GA[1])
temp_int <- HPDinterval(mcmc(sampled_variance_GA2_null[,2:8]),prob=.95)
arrows((1:7),temp_int[,1],(1:7),temp_int[,2],length=.05,code=3,angle=90,col="orange",cex=1.2)
points(c(1:7),eigen(VCV_mat_NaCl[[3]]$G1_mat/2)$values ,pch=21,bg=col_GA[2])
temp_int <- HPDinterval(mcmc(sampled_variance_GA1_null[,2:8]),prob=.95)
arrows((1:7)+.15,temp_int[,1],(1:7)+.15,temp_int[,2],length=.05,code=3,angle=90,col="orange",cex=1.2)
points(c(1:7)+.15,eigen(VCV_mat_NaCl[[4]]$G1_mat/2)$values ,pch=21,bg=col_GA[3])
legend(3,.9,"Posterior mode",bty="n")
legend(3,.6,c("GA150","GA250","GA450"),bty="n",ncol=2,pch=21,pt.bg=col_GA,col="black")
legend(3,.3,lwd=2,"95% CI of null posterior\nmodes",bty="n",col="orange")
dev.off()
pdf(file='plots/Figure7_figure_supplement3.pdf')
plot(eigen(VCV_mat_NGM[[1]]$G1_mat/2)$values~c(1:7),pch=16,xlim=c(1,7.5),bty="n",ylim=c(0.001,.83),ylab='Genetic variance',type="n",xaxt="n",xlab="",log="y")
axis(1,at=1:7,c(expression(g[max]),expression(g[2]),expression(g[3]),expression(g[4]),expression(g[5]),expression(g[6]),expression(g[7])),las=1)
temp_int <- HPDinterval(mcmc(sampled_variance_GA1_null_v2),prob=.95)
arrows((1:7)-.15,temp_int[,1],(1:7)-.15,temp_int[,2],length=.05,code=3,angle=90,col="orange",cex=1.2)
points(c(1:7)-.15,true_GA1_v2 ,pch=21,bg=col_GA[1])
temp_int <- HPDinterval(mcmc(sampled_variance_GA2_null_v2),prob=.95)
arrows((1:7),temp_int[,1],(1:7),temp_int[,2],length=.05,code=3,angle=90,col="orange",cex=1.2)
points(c(1:7),true_GA2_v2 ,pch=21,bg=col_GA[2])
temp_int <- HPDinterval(mcmc(sampled_variance_GA1_null_v2),prob=.95)
arrows((1:7)+.15,temp_int[,1],(1:7)+.15,temp_int[,2],length=.05,code=3,angle=90,col="orange",cex=1.2)
points(c(1:7)+.15,true_GA4_v2 ,pch=21,bg=col_GA[3])
points( (c(1:7)+0.2),eigen(VCV_mat_NaCl[[1]]$G1_mat/2)$values ,pch=21,bg="grey")
legend(3,.9,"Posterior mode",bty="n")
legend(3,.6,c("GA150","GA250","GA450","A6140"),bty="n",ncol=2,pch=21,pt.bg=c(col_GA,'gray'),col="black")
legend(3,.3,lwd=2,"95% CI of null posterior\nmodes",bty="n",col="orange")
dev.off()