-
Notifications
You must be signed in to change notification settings - Fork 0
/
CGtraitMF_lmer_func.R
207 lines (161 loc) · 9.68 KB
/
CGtraitMF_lmer_func.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
####Common Garden trait analysis####FOR MAT FX
#using lmer, REML mixed models#
library(lme4.0)
# or maybe?
# libary(lme4)
#with Origin and latitude as fixed effects, population and Mom as random effects#
#custom functions
#####function######
#Origin + Latitude##
CGtrait.LR<- function(trait,df,family=gaussian){
modeldata<-df[!is.na(df[[trait]]),]
modeldata$blank <- as.factor(rep("A",times=nrow(modeldata)))
modeldata$Mom<-as.factor(modeldata$Mom)
#browser()
model1<-lmer(modeldata[[trait]] ~ Origin+ Latitude +(1|PopID/Mom), family=family,data=modeldata)
model2<-lmer(modeldata[[trait]] ~ Origin+ Latitude + (1|PopID), family=family,data=modeldata) # Removes maternal family variance to test if it is a significant random effect
model3<-lmer(modeldata[[trait]] ~ Origin+ Latitude + (1|blank), family=family,data=modeldata) # Test population effect
a1 <- anova(model2,model1) # Mom is sig!
a2 <- anova(model3,model2) # pop is sig. If it says there are 0 d.f. then what you want to do is a Chi-square test using the X2 value and 1 d.f. freedom to get the p value.
modelL<-lmer(modeldata[[trait]] ~ Origin + (1|PopID), family=family,data=modeldata)
a3 <- anova(modelL, model2)
modelO<-lmer(modeldata[[trait]] ~ Latitude +(1|PopID), family=family,data=modeldata)
a4 <- anova(modelO,model2) #test for significance of origin - origin only marginally sig....!
aovs <- list(a1,a2,a3,a4)
names(aovs) <- c(paste(trait,"a1"), paste(trait,"a2"),paste(trait,"a3"),paste(trait, "a4"))
models <- list(model1,model2,model3,modelL,modelO)
names(models) <- c("model1","model2","model3","modelL","modelO")
print(aovs)
return(aovs)
}
#test for one trait, one df, specify non default family
lfcountLR<- CGtrait.LR("LfCount1",al, family=poisson)
#test for one trait, one df
shootLR<- CGtrait.LR("ShootMass.gA",al, family=gaussian)
#return models
CGtrait.models <- function(trait, df,family=gaussian){
modeldata<-df[!is.na(df[[trait]]),]
modeldata$blank <- as.factor(rep("A",times=nrow(modeldata)))
modeldata$Mom<-as.factor(modeldata$Mom)
#browser()
model1<-lmer(modeldata[[trait]] ~ Origin+ Latitude +(1|PopID/Mom), family=family,data=modeldata)
model2<-lmer(modeldata[[trait]] ~ Origin+ Latitude + (1|PopID), family=family,data=modeldata) # Removes maternal family variance to test if it is a significant random effect
model3<-lmer(modeldata[[trait]] ~ Origin+ Latitude + (1|blank), family=family,data=modeldata) # Test population effect
a1 <- anova(model2,model1) # Mom is sig!
a2 <- anova(model3,model2) # pop is sig. If it says there are 0 d.f. then what you want to do is a Chi-square test using the X2 value and 1 d.f. freedom to get the p value.
modelL<-lmer(modeldata[[trait]] ~ Origin + (1|PopID), family=family,data=modeldata)
a3 <- anova(modelL, model2)
modelO<-lmer(modeldata[[trait]] ~ Latitude +(1|PopID), family=family,data=modeldata)
a4 <- anova(modelO,model2) #test for significance of origin - origin only marginally sig....!
aovs <- list(a1,a2,a3,a4)
names(aovs) <- c(paste(trait,"a1"), paste(trait,"a2"),paste(trait,"a3"),paste(trait, "a4"))
models <- list(model1,model2,model3,modelL,modelO)
names(models) <- c("model1","model2","model3","modelL","modelO")
return(models)
}
#test for one trait, one df
shootmod <- CGtrait.models("ShootMass.gA", al) #test one trait
#for all traits in a df
#make sure all traits analyzed this way are the same distribution
names(al)#find col numbers for traits of interestes
alLR <- lapply(names(al)[8:13],function(n) CGtrait.LR(n,al))#apply func to all things in list
names(alLR) <- names(al)[8:13]
almodels <- lapply(names(al)[8:13],function(n) CGtrait.models(n,al))#apply func to all things in list
names(almodels) <- names(al)[8:13]
#Origin * Latitude#
CGtrait.LR.int<- function(trait,df,family=gaussian){
modeldata<-df[!is.na(df[[trait]]),]
modeldata$blank <- as.factor(rep("A",times=nrow(modeldata)))
modeldata$Mom<-as.factor(modeldata$Mom)
#browser()
model1<-lmer(modeldata[[trait]] ~ Origin * Latitude +(1|PopID/Mom), family=family,data=modeldata)
model2<-lmer(modeldata[[trait]] ~ Origin * Latitude + (1|PopID), family=family,data=modeldata) # Removes maternal family variance to test if it is a significant random effect
model3<-lmer(modeldata[[trait]] ~ Origin * Latitude + (1|blank), family=family,data=modeldata) # Test population effect
a1 <- anova(model2,model1) # Mom is sig!
a2 <- anova(model3,model2) # pop is sig. If it says there are 0 d.f. then what you want to do is a Chi-square test using the X2 value and 1 d.f. freedom to get the p value.
modelI <- lmer(modeldata[[trait]] ~ Origin + Latitude + (1|PopID), family=family,data=modeldata)
a3 <- anova(modelI,model2)
modelL<-lmer(modeldata[[trait]] ~ Origin + (1|PopID), family=family,data=modeldata)
a4 <- anova(modelL, modelI)
modelO<-lmer(modeldata[[trait]] ~ Latitude +(1|PopID), family=family,data=modeldata)
a5 <- anova(modelO,modelI) #test for significance of origin - origin only marginally sig....!
aovs <- list(a1,a2,a3,a4,a5)
names(aovs) <- c(paste(trait,"a1"), paste(trait,"a2"),paste(trait,"a3"),paste(trait, "a4"),paste(trait, "a5"))
models <- list(model1,model2,model3,modelI,modelL,modelO)
names(models) <- c("model1","model2","model3","modelI","modelL","modelO")
print(aovs)
return(aovs)
}
#return models
CGtrait.models.int <- function(trait, df,family=gaussian){
modeldata<-df[!is.na(df[[trait]]),]
modeldata$blank <- as.factor(rep("A",times=nrow(modeldata)))
modeldata$Mom<-as.factor(modeldata$Mom)
#browser()
model1<-lmer(modeldata[[trait]] ~ Origin * Latitude +(1|PopID/Mom), family=family,data=modeldata)
model2<-lmer(modeldata[[trait]] ~ Origin * Latitude + (1|PopID), family=family,data=modeldata) # Removes maternal family variance to test if it is a significant random effect
model3<-lmer(modeldata[[trait]] ~ Origin * Latitude + (1|blank), family=family,data=modeldata) # Test population effect
a1 <- anova(model2,model1) # Mom is sig!
a2 <- anova(model3,model2) # pop is sig. If it says there are 0 d.f. then what you want to do is a Chi-square test using the X2 value and 1 d.f. freedom to get the p value.
modelI <- lmer(modeldata[[trait]] ~ Origin + Latitude + (1|PopID), family=family,data=modeldata)
a3 <- anova(modelI,model2)
modelL<-lmer(modeldata[[trait]] ~ Origin + (1|PopID), family=family,data=modeldata)
a4 <- anova(modelL, modelI)
modelO<-lmer(modeldata[[trait]] ~ Latitude +(1|PopID), family=family,data=modeldata)
a5 <- anova(modelO,modelI) #test for significance of origin - origin only marginally sig....!
aovs <- list(a1,a2,a3,a4,a5)
names(aovs) <- c(paste(trait,"a1"), paste(trait,"a2"),paste(trait,"a3"),paste(trait, "a4"),paste(trait, "a5"))
models <- list(model1,model2,model3,modelI,modelL,modelO)
names(models) <- c("model1","model2","model3","modelI","modelL","modelO")
return(models)
}
#Origin (no latitude)#
CGtrait.LR.O<- function(trait,df,family=gaussian){
modeldata<-df[!is.na(df[[trait]]),]
modeldata$blank <- as.factor(rep("A",times=nrow(modeldata)))
modeldata$Mom<-as.factor(modeldata$Mom)
#browser()
model1<-lmer(modeldata[[trait]] ~ Origin +(1|PopID/Mom), family=family,data=modeldata)
model2<-lmer(modeldata[[trait]] ~ Origin + (1|PopID), family=family,data=modeldata) # Removes maternal family variance to test if it is a significant random effect
model3<-lmer(modeldata[[trait]] ~ Origin + (1|blank), family=family,data=modeldata) # Test population effect
a1 <- anova(model2,model1) # Mom is sig!
a2 <- anova(model3,model2) # pop is sig. If it says there are 0 d.f. then what you want to do is a Chi-square test using the X2 value and 1 d.f. freedom to get the p value.
modelO<-lmer(modeldata[[trait]] ~ (1|PopID), family=family,data=modeldata)
a3 <- anova(modelO,model2) #test for significance of origin - origin only marginally sig....!
aovs <- list(a1,a2,a3)
names(aovs) <- c(paste(trait,"a1"), paste(trait,"a2"),paste(trait,"a3"))
models <- list(model1,model2,model3,modelO)
names(models) <- c("model1","model2","model3","modelO")
print(aovs)
return(aovs)
}
#return models
CGtrait.models.O <- function(trait, df,family=gaussian){
modeldata<-df[!is.na(df[[trait]]),]
modeldata$blank <- as.factor(rep("A",times=nrow(modeldata)))
modeldata$Mom<-as.factor(modeldata$Mom)
#browser()
model1<-lmer(modeldata[[trait]] ~ Origin +(1|PopID/Mom), family=family,data=modeldata)
model2<-lmer(modeldata[[trait]] ~ Origin + (1|PopID), family=family,data=modeldata) # Removes maternal family variance to test if it is a significant random effect
model3<-lmer(modeldata[[trait]] ~ Origin + (1|blank), family=family,data=modeldata) # Test population effect
a1 <- anova(model2,model1) # Mom is sig!
a2 <- anova(model3,model2) # pop is sig. If it says there are 0 d.f. then what you want to do is a Chi-square test using the X2 value and 1 d.f. freedom to get the p value.
modelO<-lmer(modeldata[[trait]] ~ (1|PopID), family=family,data=modeldata)
a3 <- anova(modelO,model2) #test for significance of origin - origin only marginally sig....!
aovs <- list(a1,a2,a3)
names(aovs) <- c(paste(trait,"a1"), paste(trait,"a2"),paste(trait,"a3"))
models <- list(model1,model2,model3,modelO)
names(models) <- c("model1","model2","model3","modelO")
return(models)
}
###########normality???#####
#to get one model
almodels[[1]][1] #first number is trait in column order of df, second number is model number
names(almodels[1]) #to verify trait
#to check normality of residuals
cuRoot.lmer <- cumodels$model2
plot(resid(nmodels[2]) ~ fitted(nmodels[2]),main="residual plot")
abline(h=0)
# checking the normality of residuals e_i:
qqnorm(resid(nRootlog.lmer), main="Q-Q plot for residuals")
qqline(resid(nRootlog.lmer))