Higher parasite burdens are associated with lower body condition in a Norwegian moose population
Jaelle C. Brealey
October 20, 2025
Abstract
Parasites are ubiquitous in wildlife and can influence health and fitness. Thus, the inclusion of parasites in wildlife monitoring programs could provide important information for understanding declines in population body condition or recruitment rates. However, parasites are often neglected as a factor in wildlife management. We characterised the gastrointestinal parasite community of moose (Alces alces) in central Norway over an autumn hunting season and the following winter, using parasitological examination of faecal samples. We identified an age-related effect on parasite communities, including increased prevalence and abundance of Protostrongylidae species and increased community richness in yearlings compared to older moose. We also found that individuals with a lower body mass had higher burdens of Trichostrongylidae and Protostrongylidae species in autumn. There was some evidence for increased parasite burdens in males compared to females, while there was less evidence for seasonal differences between autumn and winter in a subsample of adult females. We found no strong associations between parasite burdens and moose population density. Our results add to the growing body of literature supporting the use of parasite burdens as proxies for an individual’s health status and potentially as predictors for fitness effects. In many Fennoscandian moose populations, the body condition and recruitment rates of individuals are declining, thus further monitoring of parasites may be needed to better understand the mechanisms behind these declines, including the potential role of parasites.
Here we provide the code used for the analysis of this manuscript. The necessary dataset is available as Supplementary Information from the journal website.
Functions
Custom functions required for the analysis.
run_glm <- function(df, resp_vars, covar_formula, fam, ct_round = FALSE, min.samples = 5){
# set up empty data frame for summary results
res <- data.frame(
resp_var=NULL,
covariates=NULL,
estimate=NULL,
std.error=NULL,
statistic=NULL,
p.value=NULL
)
# set up empty list to hold models
mods <- list()
if(fam == "binomial"){
# run logistic GLM for each response variable
for(x in resp_vars){
# only run if there are enough positive samples
if(sum(df[,x], na.rm = T) >= min.samples){
print(paste0("Running logistic model for ",x)) # include to know which variables trigger warnings
f <- paste0(x,covar_formula)
mod <- glm(formula = formula(f),
data = df,
family = binomial(link = "logit"),
na.action = na.omit) # remove NAs
mod.sum <- summary(mod)
colnames(mod.sum$coefficients) <- c("estimate","std.error","statistic","p.value")
res <- rbind.data.frame(
res,
data.frame(
resp_var=x,
covariates=rownames(mod.sum$coefficients),
as.data.frame(mod.sum$coefficients),
row.names = NULL
)
)
mods[[x]] <- mod
} else {
print(paste0("Skipping ",x,", not enough positive samples."))
next
}
}
} else if(fam == "gaussian"){
# run normal GLM for each response variable
for(x in resp_vars){
# only run if there are enough positive samples
if(sum(df[,x] > 0, na.rm = T) >= min.samples){
print(paste0("Running gaussian model for ",x)) # include to know which variables trigger warnings
# apply pseudo-count of 1 and log-transform
df$y_trans <- log10(df[,x] + 1)
f <- paste0("y_trans",covar_formula)
mod <- glm(formula = formula(f),
data = df,
family = gaussian,
na.action = na.omit) # remove NAs
mod.sum <- summary(mod)
colnames(mod.sum$coefficients) <- c("estimate","std.error","statistic","p.value")
res <- rbind.data.frame(
res,
data.frame(
resp_var=x,
covariates=rownames(mod.sum$coefficients),
as.data.frame(mod.sum$coefficients),
row.names = NULL
)
)
mods[[x]] <- mod
} else {
print(paste0("Skipping ",x,", not enough positive samples."))
next
}
}
} else if(fam == "poisson"){
# run poisson GLM for each response variable
for(x in resp_vars){
# only run if there are enough positive samples
if(sum(df[,x] > 0, na.rm = T) >= min.samples){
print(paste0("Running Poisson model for ",x)) # include to know which variables trigger warnings
if(ct_round){
# round numeric data to integers (note may not always be valid!)
df$y_rounded <- round(df[,x], digits = 0)
f <- paste0("y_rounded",covar_formula)
} else {
f <- paste0(x,covar_formula)
}
mod <- glm(formula = formula(f),
data = df,
family = poisson(link = "log"),
na.action = na.omit) # remove NAs
# check for overdispersion
dis <- performance::check_overdispersion(mod)
if(dis$p_value < 0.05){
# overdispersion detected, run quasi-poisson
print(paste0("Overdispersion detected; running quasi-Poisson model for ",x))
mod <- update(mod, family=quasipoisson)
}
mod.sum <- summary(mod)
colnames(mod.sum$coefficients) <- c("estimate","std.error","statistic","p.value")
res <- rbind.data.frame(
res,
data.frame(
resp_var=x,
covariates=rownames(mod.sum$coefficients),
as.data.frame(mod.sum$coefficients),
row.names = NULL
)
)
mods[[x]] <- mod
} else {
print(paste0("Skipping ",x,", not enough positive samples."))
next
}
}
} else if(fam %in% c("nbinom1","nbinom2")) {
# a generalised form of Poisson that deals with overdispersion
# use glmmTMB library
for(x in resp_vars){
# only run if there are enough positive samples
if(sum(df[,x] > 0, na.rm = T) >= min.samples){
print(paste0("Running Neg Binomial model for ",x)) # include to know which variables trigger warnings
if(ct_round){
# round numeric data to integers (note may not always be valid!)
df$y_rounded <- round(df[,x], digits = 0)
f <- paste0("y_rounded",covar_formula)
} else {
f <- paste0(x,covar_formula)
}
if(fam == "nbinom1"){
# for linear increase in variance with increasing mean
mod <- glmmTMB(formula = formula(f),
data = df,
family = nbinom1,
na.action = na.omit) # remove NAs
} else if(fam == "nbinom2"){
# for quadratic (rapid) increase in variance with increasing mean
mod <- glmmTMB(formula = formula(f),
data = df,
family = nbinom2,
na.action = na.omit) # remove NAs
} else {
print("Family not implemented yet.")
break
}
mod.sum <- summary(mod)
colnames(mod.sum$coefficients$cond) <- c("estimate","std.error","statistic","p.value")
res <- rbind.data.frame(
res,
data.frame(
resp_var=x,
covariates=rownames(mod.sum$coefficients$cond),
as.data.frame(mod.sum$coefficients$cond),
row.names = NULL
)
)
mods[[x]] <- mod
} else {
print(paste0("Skipping ",x,", not enough positive samples."))
next
}
}
} else {
print("Family not implemented yet.")
break
}
# if enough tests, run BH correction
if(length(unique(res$resp_var)) > 1){
res$p.adj <- p.adjust(res$p.value,"BH")
}
return(list(
summary=res,
models=mods))
}
calc_median_IQR_label <- function(i, v, groups,
sum.tab = ms.tab1, df.meta = TA16.sub,
intensity = FALSE){
# default: calculate abundance (all samples, including 0s)
# option: calculate intensity (only positive (>0) samples)
if(intensity){
df.meta <- df.meta[which(df.meta[,v] > 0),]
if(nrow(df.meta) < 3){
return(NA)
}
}
if("all" %in% as.character(unlist(sum.tab[i,groups]))){
# all samples
m <- median(df.meta[,v], na.rm = T)
q1 <- quantile(df.meta[,v], probs = 0.25, na.rm = T)[[1]]
q3 <- quantile(df.meta[,v], probs = 0.75, na.rm = T)[[1]]
} else {
# loop through and subset by groups
for(g in groups){
gg <- as.character(sum.tab[i,g])
df.meta <- df.meta[which(df.meta[,g] == gg),]
}
if(nrow(df.meta) < 3){
# if less than 3 samples in the group, can't calculate median
return(NA)
} else {
m <- median(df.meta[,v], na.rm = T)
q1 <- quantile(df.meta[,v], probs = 0.25, na.rm = T)[[1]]
q3 <- quantile(df.meta[,v], probs = 0.75, na.rm = T)[[1]]
}
}
l <- paste0(round(m, digits = 0)," (",
round(q1, digits = 0),"-",
round(q3, digits = 0),")")
}
format_rich_summary <- function(mod){
tab.sum <- summary(mod)
colnames(tab.sum$coefficients) <- c("estimate","std.error","statistic","p.value")
tab2.sum <- data.frame(
resp_var="num.of.parasite.groups",
covariates=row.names(tab.sum$coefficients),
as.data.frame(tab.sum$coefficients),
p.adj=NA,
row.names = NULL
)
return(tab2.sum)
}
format_posthoc_summary <- function(mod, resp_var_name){
# for "summary.glht" class
tab.sum <- summary(mod)
tab2.sum <- data.frame(
resp_var=resp_var_name,
covariates=names(tab.sum$test$coefficients),
estimate=tab.sum$test$coefficients,
std.error=tab.sum$test$sigma,
statistic=tab.sum$test$tstat,
p.value=tab.sum$test$pvalues,
p.adj=NA,
row.names = NULL
)
return(tab2.sum)
}Load data
Data was generated through parasitological examination of faecal samples, using a modified McMasters method to estimate presence and abundance of endoparasitic eggs and oocysts and the Baermann technique quantify and identify parasitic L1 stage larvae in the faeces. The dataset used here is available from Supplementary Information (Table S5).
# set encoding to latin1, to handle additional Norwegian vowels
dat <- read.delim("figures/supp_dataset.txt", sep = "\t", fileEncoding = "latin1")
# set up levels for factors
dat$Season <- factor(dat$Season)
dat$Sex <- factor(dat$Sex,
levels = c("Male","Female"))
dat$Adult_age_category <- factor(dat$Adult_age_category,
levels = c("agegroup1","agegroup2","agegroup3"),
ordered = T)
dat$Carcass_mass_category <- factor(dat$Carcass_mass_category,
levels = c("Normal","Low"),
ordered = T)
dat$Municipality <- factor(dat$Municipality,
ordered = F)
# add days spent in storage and post as potential confounders
dat$Days.harvested.vs.posted <- as.integer(as.Date(dat$Date_posted, format = "%d.%m.%Y") - as.Date(dat$Date_harvested, format = "%d.%m.%Y"))
dat$Days.harvested.vs.lab <- as.integer(as.Date(dat$Date_received_in_lab, format = "%d.%m.%Y") - as.Date(dat$Date_harvested, format = "%d.%m.%Y"))
dat$Days.posted.vs.lab <- as.integer(as.Date(dat$Date_received_in_lab, format = "%d.%m.%Y") - as.Date(dat$Date_posted, format = "%d.%m.%Y"))
# Parasite groups
mcmaster.parasite.groups.01 <- c("Eimeria01", "Strongylidetypeegg01", "NematodirusNematodirellaEgg01",
"Trichuris01", "Capillaria01", "MiddEgg01",
"Moniezia01", "Strongyloides01")
baermann.parasite.groups.01 <- c("ProtostrongylideLarver01", "EalcesLarvae01",
"ValcesLarvae01", "DictyocaulusLarver01")
mcmaster.parasite.groups.EPG <- c("EimeriaOPG", "StrongylidetypeeggEPG", "NematodirussppEPG", "NematodirusBattusEPG",
"TrichurisEPG", "CapillariaEPG", "MiddEggEPG")
baermann.parasite.groups.LPG <- c("ProtostrongylideLarverLPG", "DictyocaulusLPG")Autumn 2016 Trøndelag
First, focus on associations within the group of hunted adult moose (1+ years) from autumn 2016 (code TA16).
Summaries
Generate some basic summaries of metadata.
# subset to Trøndelag in 2016
TA16 <- dat[which(dat$Year == "2016" & dat$Season == "Autumn"),]
# exclude samples missing key data
TA16.sub <- TA16[which(!is.na(TA16$Sex) & !is.na(TA16$Adult_age_category) & !is.na(TA16$Carcass_mass_category) & !is.na(TA16$Municipality)),]Fig. S1
ggplot(TA16.sub, aes(Carcass_mass_category, Carcass_mass_kg))+
geom_boxplot(aes(fill = Sex), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Sex), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c(Male="#E41A1C",Female="#377EB8"), labels = c(Male="Male", Female="Female"))+
labs(x = "Carcass mass category", y = "Carcass mass (kg)", fill = "Sex")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "right",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ Adult_age_category, nrow = 1,
labeller = labeller(Adult_age_category=c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr")))
ggsave("figures/FigS1_Mass_by_SexMassCat.png",units = "in", dpi = 1200, width = 8, height = 4)
# ggsave("figures/Fig2.eps",units = "in", dpi = 1200, width = 8, height = 4)Other figures
ggplot(TA16.sub, aes(Adult_age_category, Carcass_mass_kg))+
geom_boxplot(aes(fill = Sex), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Sex), shape = 21, color = "black", size = 2,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#E41A1C","#377EB8"))+
scale_x_discrete(labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
labs(x = "Age group", y = "Carcass mass (kg)", fill = "Sex")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
ggplot(TA16.sub, aes(Carcass_mass_category, Carcass_mass_kg))+
geom_boxplot(aes(fill = Carcass_mass_category), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 2,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
labs(x = "Weight category", y = "Carcass mass (kg)")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "none",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ Adult_age_category, nrow = 1,
labeller = labeller(Adult_age_category=c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr")))
ggplot(TA16.sub, aes(Municipality, MunicipalityDensity_2016))+
geom_boxplot(aes(fill = Municipality), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 2,
position = position_jitterdodge(jitter.width = 0.2))+
labs(x = "Municipality", y = "Harvest density (2016)")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "none",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
ggplot(TA16.sub, aes(Municipality, MunicipalityDensity_Mean2016_18))+
geom_boxplot(aes(fill = Municipality), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 2,
position = position_jitterdodge(jitter.width = 0.2))+
labs(x = "Municipality", y = "Harvest density (mean 2016-2018)")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "none",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
ggplot(TA16.sub, aes(MunicipalityDensity_2016, Carcass_mass_kg))+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 2)+
# scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
labs(x = "Harvest density (2016)", y = "Carcass mass (kg)")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
ggplot(TA16.sub, aes(MunicipalityDensity_Mean2016_18, Carcass_mass_kg))+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 2)+
# scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
labs(x = "Harvest density (2016-2018)", y = "Carcass mass (kg)")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
Table 1a
Basic stats
table(TA16.sub$McMaster_performed)##
## 1
## 117table(TA16.sub$Baermann_performed)##
## 0 1
## 14 103table(TA16.sub$Sex, TA16.sub$Adult_age_category)##
## agegroup1 agegroup2 agegroup3
## Male 46 12 1
## Female 29 17 12sum(table(TA16.sub$Sex, TA16.sub$Adult_age_category, TA16.sub$Carcass_mass_category))## [1] 117table(TA16.sub$Carcass_mass_category, TA16.sub$Adult_age_category)##
## agegroup1 agegroup2 agegroup3
## Normal 58 23 10
## Low 17 6 3prop.table(table(TA16.sub$Carcass_mass_category, TA16.sub$Adult_age_category), margin = 2)##
## agegroup1 agegroup2 agegroup3
## Normal 0.7733333 0.7931034 0.7692308
## Low 0.2266667 0.2068966 0.2307692summary(TA16.sub[which(TA16.sub$Adult_age_category == "agegroup1" & TA16.sub$Carcass_mass_category == "Normal"),"Carcass_mass_kg"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 107.0 120.0 127.5 130.0 137.8 196.0summary(TA16.sub[which(TA16.sub$Adult_age_category == "agegroup1" & TA16.sub$Carcass_mass_category == "Low"),"Carcass_mass_kg"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 81.00 96.00 104.00 99.82 105.00 106.00summary(TA16.sub[which(TA16.sub$Adult_age_category == "agegroup2" & TA16.sub$Carcass_mass_category == "Normal"),"Carcass_mass_kg"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 140.0 146.5 155.0 158.6 167.0 217.0summary(TA16.sub[which(TA16.sub$Adult_age_category == "agegroup2" & TA16.sub$Carcass_mass_category == "Low"),"Carcass_mass_kg"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 113.0 118.2 120.5 124.3 132.5 138.0summary(TA16.sub[which(TA16.sub$Adult_age_category == "agegroup3" & TA16.sub$Carcass_mass_category == "Normal"),"Carcass_mass_kg"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 160.0 181.5 199.5 197.1 208.0 240.0summary(TA16.sub[which(TA16.sub$Adult_age_category == "agegroup3" & TA16.sub$Carcass_mass_category == "Low"),"Carcass_mass_kg"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 145.0 145.5 146.0 149.3 151.5 157.0# summary table
ms.tab1 <- as.data.frame(table(TA16.sub$Sex, TA16.sub$Adult_age_category, TA16.sub$Carcass_mass_category))
names(ms.tab1) <- c("Sex","Adult_age_category","Carcass_mass_category","count")
ms.tab1 <- rbind.data.frame(
ms.tab1,
data.frame(Sex="all", Adult_age_category="all", Carcass_mass_category="all", count=nrow(TA16.sub))
)
ms.tab1$Carcass_mass_kg.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "Carcass_mass_kg",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub)
# add major parasite groups
# same as abundance/prevalence models, must be at least 5 positive samples to include in table
# v1 - intensity
ms.tab1$StrongylidetypeeggEPG.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "StrongylidetypeeggEPG",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub, intensity = TRUE)
ms.tab1$NematodirussppEPG.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "NematodirussppEPG",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub, intensity = TRUE)
ms.tab1$NematodirusBattusEPG.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "NematodirusBattusEPG",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub, intensity = TRUE)
ms.tab1$TrichurisEPG.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "TrichurisEPG",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub, intensity = TRUE)
ms.tab1$CapillariaEPG.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "CapillariaEPG",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub, intensity = TRUE)
ms.tab1$EimeriaOPG.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "EimeriaOPG",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub, intensity = TRUE)
ms.tab1$ProtostrongylideLarverLPG.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "ProtostrongylideLarverLPG",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub, intensity = TRUE)
ms.tab1$DictyocaulusLPG.Median <- sapply(seq(1,nrow(ms.tab1)), calc_median_IQR_label, v = "DictyocaulusLPG",
groups = c("Sex","Adult_age_category","Carcass_mass_category"),
sum.tab = ms.tab1, df.meta = TA16.sub, intensity = TRUE)
ms.tab1$Season <- "Autumn"Generate some basic prevalence and abundance plots.
TA16.abund.df <- melt(
TA16.sub[,c("IndividualID","Sex","Adult_age_category","Carcass_mass_kg","Exact_age","Carcass_mass_category",
"Municipality","MunicipalityDensity_2016","MunicipalityDensity_Mean2016_18","McMaster_performed","Baermann_performed",
mcmaster.parasite.groups.EPG, baermann.parasite.groups.LPG)],
id.vars = c("IndividualID","Sex","Adult_age_category","Carcass_mass_kg","Exact_age","Carcass_mass_category","Municipality","MunicipalityDensity_2016","MunicipalityDensity_Mean2016_18","McMaster_performed","Baermann_performed"),
measure.vars = c(mcmaster.parasite.groups.EPG, baermann.parasite.groups.LPG)
)ggplot(subset(TA16.abund.df, !variable %in% c("MiddEggEPG","NematodirusBattusEPG")),
aes(Adult_age_category, value))+
geom_boxplot(aes(fill = Sex), alpha = 0.25, outlier.colour = NA)+
geom_point(aes(fill = Sex), shape = 21, color = "black", size = 2,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#E41A1C","#377EB8"))+
labs(x = "Age group", y = "Abundance", fill = "Sex")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable, scales = "free_y", nrow = 2)## Warning: Removed 34 rows containing non-finite outside the scale range
## (`stat_boxplot()`).## Warning: Removed 34 rows containing missing values or values outside the scale range
## (`geom_point()`).
# exclude parasites with < 5 detections
ggplot(subset(TA16.abund.df, !variable %in% c("MiddEggEPG","NematodirusBattusEPG") & !is.na(Carcass_mass_kg)),
aes(Carcass_mass_kg, value))+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 2)+
scale_y_continuous(transform = "log10")+
scale_fill_manual(values = c("#C7E9C0","#74C476","#006D2C"))+
labs(x = "Carcass mass (kg)", y = "Abundance", fill = "Age group")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable, scales = "free_y", nrow = 2)## Warning in scale_y_continuous(transform = "log10"): log-10 transformation
## introduced infinite values.## Warning: Removed 34 rows containing missing values or values outside the scale range
## (`geom_point()`).
ggplot(subset(TA16.abund.df, !variable %in% c("MiddEggEPG","NematodirusBattusEPG")), aes(Adult_age_category, value))+
geom_boxplot(aes(fill = Carcass_mass_category), alpha = 0.25, outlier.colour = NA)+
geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 2,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
scale_x_discrete(labels = c("1 yr","2-5 yr","6+ yr"))+
labs(x = "Age category", y = "Abundance", fill = "Weight category")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable, scales = "free_y", nrow = 2)## Warning: Removed 34 rows containing non-finite outside the scale range
## (`stat_boxplot()`).## Warning: Removed 34 rows containing missing values or values outside the scale range
## (`geom_point()`).
Table 2a
Autumn, summary table of prevalence by age and sex.
TA16.det.df <- melt(
TA16.sub[,c("IndividualID","Sex","Adult_age_category","Carcass_mass_kg","Exact_age","Carcass_mass_category",
"Municipality","MunicipalityDensity_2016","MunicipalityDensity_Mean2016_18","McMaster_performed","Baermann_performed",
mcmaster.parasite.groups.01, baermann.parasite.groups.01)],
id.vars = c("IndividualID","Sex","Adult_age_category","Carcass_mass_kg","Exact_age","Carcass_mass_category","Municipality","MunicipalityDensity_2016","MunicipalityDensity_Mean2016_18","McMaster_performed","Baermann_performed"),
measure.vars = c(mcmaster.parasite.groups.01, baermann.parasite.groups.01)
)
TA16.det.by.group <- dcast(
TA16.det.df,
Sex + Adult_age_category ~ variable,
value.var = "value",
fun.aggregate = sum, fill = 0, drop = T, na.rm = T
)
## Calculate no. of individuals based on samples for which test was performed
TA16.det.by.group$num_inds.MM <- sapply(seq(1,nrow(TA16.det.by.group)), function(i){
sex <- TA16.det.by.group[i,"Sex"]
age <- TA16.det.by.group[i,"Adult_age_category"]
length(which(TA16.sub$Sex == sex & TA16.sub$Adult_age_category == age & TA16.sub$McMaster_performed == 1))
})
TA16.det.by.group$num_inds.BA <- sapply(seq(1,nrow(TA16.det.by.group)), function(i){
sex <- TA16.det.by.group[i,"Sex"]
age <- TA16.det.by.group[i,"Adult_age_category"]
length(which(TA16.sub$Sex == sex & TA16.sub$Adult_age_category == age & TA16.sub$Baermann_performed == 1))
})
TA16.det.by.group.hm <- melt(TA16.det.by.group,
id.vars = c("Sex","Adult_age_category","num_inds.MM","num_inds.BA"),
measure.vars = c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
variable.name = "parasite", value.name = "n.positive")
TA16.det.by.group.hm$group <- paste(TA16.det.by.group.hm$Adult_age_category, TA16.det.by.group.hm$Sex, sep = "_")
TA16.det.by.group.hm$proportion <- ifelse(
TA16.det.by.group.hm$parasite %in% mcmaster.parasite.groups.01,
TA16.det.by.group.hm$n.positive / TA16.det.by.group.hm$num_inds.MM,
TA16.det.by.group.hm$n.positive / TA16.det.by.group.hm$num_inds.BA
)
# add 95%CI to prevalence
TA16.det.by.group.hm$CI95.lower <- sapply(seq(1,nrow(TA16.det.by.group.hm)), function(i){
p <- as.character(TA16.det.by.group.hm[i,"parasite"])
n.det <- TA16.det.by.group.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TA16.det.by.group.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TA16.det.by.group.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[1])
})
TA16.det.by.group.hm$CI95.upper <- sapply(seq(1,nrow(TA16.det.by.group.hm)), function(i){
p <- as.character(TA16.det.by.group.hm[i,"parasite"])
n.det <- TA16.det.by.group.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TA16.det.by.group.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TA16.det.by.group.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[2])
})
TA16.det.by.group.hm$parasite <- factor(TA16.det.by.group.hm$parasite,
levels = c("MiddEgg01", "Moniezia01", "Eimeria01",
"Strongyloides01", "Capillaria01", "Trichuris01", "NematodirusNematodirellaEgg01",
"Strongylidetypeegg01",
"ValcesLarvae01", "EalcesLarvae01", "ProtostrongylideLarver01",
"DictyocaulusLarver01"))
# exclude mites
TA16.det.by.group.hm <- TA16.det.by.group.hm[which(TA16.det.by.group.hm$parasite != "MiddEgg01"),]
TA16.det.by.group.hm$parasite <- droplevels(TA16.det.by.group.hm$parasite)
TA16.det.by.group[,c(1:2,15:16)]## Sex Adult_age_category num_inds.MM num_inds.BA
## 1 Male agegroup1 46 40
## 2 Male agegroup2 12 11
## 3 Male agegroup3 1 1
## 4 Female agegroup1 29 27
## 5 Female agegroup2 17 14
## 6 Female agegroup3 12 10TA16.det.by.group.hm$percent <- round(TA16.det.by.group.hm$proportion * 100, 1)
TA16.det.by.group.hm$label_v2 <- sapply(seq(1,nrow(TA16.det.by.group.hm)), function(i){
ct <- TA16.det.by.group.hm[i, "n.positive"]
if(ct == 0){
return(0)
} else {
per <- TA16.det.by.group.hm[i,"percent"]
return(paste0(ct," (",per,"%)"))
}
})
TA16.det.by.group.hm$label_v3 <- sapply(seq(1,nrow(TA16.det.by.group.hm)), function(i){
if(TA16.det.by.group.hm[i,"proportion"] == 0){
return(0)
} else {
prop <- round(TA16.det.by.group.hm[i,"proportion"], digits = 2)
ciL <- round(TA16.det.by.group.hm[i,"CI95.lower"], digits = 1)
ciU <- round(TA16.det.by.group.hm[i,"CI95.upper"], digits = 1)
return(paste0(prop,"\n(",ciL,"-",ciU,")"))
}
})
TA16.det.by.group.hm$group <- factor(as.character(TA16.det.by.group.hm$group),
levels = c("agegroup1_Male", "agegroup2_Male", "agegroup3_Male",
"agegroup1_Female", "agegroup2_Female", "agegroup3_Female"))prev.fig1 <-
ggplot(TA16.det.by.group.hm, aes(group, parasite, fill = proportion))+
geom_tile(colour = "grey50", linewidth = 0.5)+
scale_fill_gradient(low="white", high="grey50")+
# geom_vline(xintercept = 3.5, color = "black", linewidth = 0.5, linetype = 2)+
geom_text(aes(label=label_v3))+
scale_y_discrete(labels = c("*Moniezia* spp.<br>eggs","*Eimeria* spp.<br>oocysts",
"*Strongyloides* spp.<br>eggs","*Capillaria* spp.<br>eggs","*Trichuris* spp.<br>eggs","Nematodirinae<br>eggs",
"Trichostrongylidae<br>spp.", "*V. alces*<br>larvae","*E. alces*<br>larvae","Protostrongylidae<br>larvae",
"*Dictyocaulus* spp.<br>larvae"))+
scale_x_discrete(labels = c("Autumn\nMale\n1 yr\nnL=40\nnE=46","Autumn\nMale\n2-5 yr\nnL=11\nnE=12","Autumn\nMale\n6+ yr\nnL=1\nnE=1",
"Autumn\nFemale\n1 yr\nnL=27\nnE=29","Autumn\nFemale\n2-5 yr\nnL=14\nnE=17","Autumn\nFemale\n6+ yr\nnL=10\nnE=12"),
position = "top")+
labs(y = "", x = "", fill = "Prevalence \n")+
# tag = "Season\nSex\nAge group\nnL\nnE")+
# annotate("text", x = -1, y = 12, label = "Season\nSex\nnL\nnE", size = 2, colour = "black")+
theme_bw()+
guides(fill = guide_colorbar(barwidth = 10, barheight = 1))+
theme(axis.title = element_text(size = 13, colour = "black"),
axis.text.y = ggtext::element_markdown(size = 12, colour = "black", hjust = 1),
axis.text.x = element_text(size = 12, colour = "black", angle = 0, vjust = 0.5),
legend.text = element_text(size = 11, colour = "black"), legend.title = element_text(size = 12, colour = "black"),
legend.position = "bottom",
panel.border = element_blank(), axis.ticks = element_blank()
# plot.tag.position = c(0.0698, 0.938), plot.tag = element_text(size = 12, colour = "black", angle = 0, hjust = 0)
)
prev.fig1
prop.table(table(TA16.sub$Strongylidetypeegg01))##
## 0 1
## 0.01709402 0.98290598Models
GLMs testing for associations between sex (factor) and age group (ordered factor), as well as carcass mass after accounting for sex and age. Includes adjustment for multiple hypothesis testing.
Prevalence
Must be present in at least 5 samples to test (otherwise not enough variation in dataset). Six models per parasite:
- exact age + sex + weight + municipality
- age group + sex + weight category + municipality
- age group + sex * weight category + municipality
- exact age * weight category + sex + municipality
- age group + sex + weight category + harvest density (2016)
- age group + sex + weight category + harvest density (mean over 2016-2018)
## MAIN: Exact Age and MunicipalityDensity_Mean2016_18 to reduce comparisons in model
TA16.prev.r2 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
" ~ Exact_age + Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "binomial",
min.samples = 5
)## [1] "Skipping Eimeria01, not enough positive samples."
## [1] "Running logistic model for Strongylidetypeegg01"
## [1] "Running logistic model for NematodirusNematodirellaEgg01"
## [1] "Running logistic model for Trichuris01"
## [1] "Skipping Capillaria01, not enough positive samples."
## [1] "Skipping MiddEgg01, not enough positive samples."
## [1] "Skipping Moniezia01, not enough positive samples."
## [1] "Running logistic model for Strongyloides01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ProtostrongylideLarver01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for EalcesLarvae01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ValcesLarvae01"
## [1] "Skipping DictyocaulusLarver01, not enough positive samples."## SUPP:
# kg continuous
TA16.prev.r1 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
" ~ Exact_age + Carcass_mass_kg + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "binomial",
min.samples = 5
)## [1] "Skipping Eimeria01, not enough positive samples."
## [1] "Running logistic model for Strongylidetypeegg01"
## [1] "Running logistic model for NematodirusNematodirellaEgg01"
## [1] "Running logistic model for Trichuris01"
## [1] "Skipping Capillaria01, not enough positive samples."
## [1] "Skipping MiddEgg01, not enough positive samples."
## [1] "Skipping Moniezia01, not enough positive samples."
## [1] "Running logistic model for Strongyloides01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ProtostrongylideLarver01"
## [1] "Running logistic model for EalcesLarvae01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ValcesLarvae01"
## [1] "Skipping DictyocaulusLarver01, not enough positive samples."# # age group
# TA16.prev.r2 <- run_glm(
# TA16.sub,
# c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
# " ~ Adult_age_category + Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18",
# fam = "binomial",
# min.samples = 5
# )
# 2016 density
TA16.prev.r3 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
" ~ Exact_age + Carcass_mass_category + Sex + MunicipalityDensity_2016 + Days.harvested.vs.lab",
fam = "binomial",
min.samples = 5
)## [1] "Skipping Eimeria01, not enough positive samples."
## [1] "Running logistic model for Strongylidetypeegg01"
## [1] "Running logistic model for NematodirusNematodirellaEgg01"
## [1] "Running logistic model for Trichuris01"
## [1] "Skipping Capillaria01, not enough positive samples."
## [1] "Skipping MiddEgg01, not enough positive samples."
## [1] "Skipping Moniezia01, not enough positive samples."
## [1] "Running logistic model for Strongyloides01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ProtostrongylideLarver01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for EalcesLarvae01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ValcesLarvae01"
## [1] "Skipping DictyocaulusLarver01, not enough positive samples."# Municipality
TA16.prev.r4 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
" ~ Exact_age + Carcass_mass_category + Sex + Municipality + Days.harvested.vs.lab",
fam = "binomial",
min.samples = 5
)## [1] "Skipping Eimeria01, not enough positive samples."
## [1] "Running logistic model for Strongylidetypeegg01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for NematodirusNematodirellaEgg01"
## [1] "Running logistic model for Trichuris01"
## [1] "Skipping Capillaria01, not enough positive samples."
## [1] "Skipping MiddEgg01, not enough positive samples."
## [1] "Skipping Moniezia01, not enough positive samples."
## [1] "Running logistic model for Strongyloides01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ProtostrongylideLarver01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for EalcesLarvae01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ValcesLarvae01"
## [1] "Skipping DictyocaulusLarver01, not enough positive samples."# sex x weight interaction term
TA16.prev.r5 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
" ~ Exact_age + Carcass_mass_category * Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "binomial",
min.samples = 5
)## [1] "Skipping Eimeria01, not enough positive samples."
## [1] "Running logistic model for Strongylidetypeegg01"
## [1] "Running logistic model for NematodirusNematodirellaEgg01"
## [1] "Running logistic model for Trichuris01"
## [1] "Skipping Capillaria01, not enough positive samples."
## [1] "Skipping MiddEgg01, not enough positive samples."
## [1] "Skipping Moniezia01, not enough positive samples."
## [1] "Running logistic model for Strongyloides01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ProtostrongylideLarver01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for EalcesLarvae01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ValcesLarvae01"
## [1] "Skipping DictyocaulusLarver01, not enough positive samples."# age x kg cat interaction term
TA16.prev.r6 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
" ~ Exact_age * Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "binomial",
min.samples = 5
)## [1] "Skipping Eimeria01, not enough positive samples."
## [1] "Running logistic model for Strongylidetypeegg01"
## [1] "Running logistic model for NematodirusNematodirellaEgg01"
## [1] "Running logistic model for Trichuris01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Skipping Capillaria01, not enough positive samples."
## [1] "Skipping MiddEgg01, not enough positive samples."
## [1] "Skipping Moniezia01, not enough positive samples."
## [1] "Running logistic model for Strongyloides01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ProtostrongylideLarver01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for EalcesLarvae01"## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## [1] "Running logistic model for ValcesLarvae01"
## [1] "Skipping DictyocaulusLarver01, not enough positive samples."TA16.prev.r1$summary[which(TA16.prev.r1$summary$covariates != "(Intercept)" & TA16.prev.r1$summary$p.value < 0.05),]## resp_var covariates estimate std.error statistic
## 26 ProtostrongylideLarver01 Exact_age -1.706975 0.5869081 -2.908420
## 32 EalcesLarvae01 Exact_age -2.597049 0.8702038 -2.984415
## 38 ValcesLarvae01 Exact_age -1.215502 0.5123512 -2.372399
## p.value p.adj
## 26 0.003632602 0.05085642
## 32 0.002841214 0.05085642
## 38 0.017672979 0.14845303TA16.prev.r2$summary[which(TA16.prev.r2$summary$covariates != "(Intercept)" & TA16.prev.r2$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 26 ProtostrongylideLarver01 Exact_age -2.2002240 0.5571874
## 32 EalcesLarvae01 Exact_age -3.3206061 0.8611806
## 33 EalcesLarvae01 Carcass_mass_category.L 1.7491054 0.8007562
## 38 ValcesLarvae01 Exact_age -1.3140766 0.4485914
## 39 ValcesLarvae01 Carcass_mass_category.L 0.9670895 0.4926710
## statistic p.value p.adj
## 26 -3.948805 7.854244e-05 0.00242163
## 32 -3.855876 1.153157e-04 0.00242163
## 33 2.184317 2.893896e-02 0.15192952
## 38 -2.929340 3.396827e-03 0.02419897
## 39 1.962952 4.965178e-02 0.20853749TA16.prev.r3$summary[which(TA16.prev.r3$summary$covariates != "(Intercept)" & TA16.prev.r3$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 26 ProtostrongylideLarver01 Exact_age -2.190376 0.5526011
## 32 EalcesLarvae01 Exact_age -3.330831 0.8629707
## 33 EalcesLarvae01 Carcass_mass_category.L 1.736644 0.8013825
## 38 ValcesLarvae01 Exact_age -1.314138 0.4454480
## statistic p.value p.adj
## 26 -3.963756 7.377956e-05 0.002383799
## 32 -3.859726 1.135142e-04 0.002383799
## 33 2.167059 3.023032e-02 0.158709174
## 38 -2.950150 3.176196e-03 0.026680043TA16.prev.r4$summary[which(TA16.prev.r4$summary$covariates != "(Intercept)" & TA16.prev.r4$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 24 Trichuris01 MunicipalitySelbu -1.893457 0.8779352
## 38 ProtostrongylideLarver01 Exact_age -2.470020 0.6432193
## 47 EalcesLarvae01 Exact_age -4.257373 1.2480872
## 48 EalcesLarvae01 Carcass_mass_category.L 1.956695 0.8870405
## 56 ValcesLarvae01 Exact_age -1.505810 0.5135315
## 57 ValcesLarvae01 Carcass_mass_category.L 1.001546 0.4983439
## 62 ValcesLarvae01 MunicipalityTydal 3.896743 1.7677963
## statistic p.value p.adj
## 24 -2.156716 0.0310278123 0.279250311
## 38 -3.840090 0.0001229893 0.007748328
## 47 -3.411118 0.0006469712 0.020379593
## 48 2.205869 0.0273931477 0.279250311
## 56 -2.932264 0.0033650101 0.070665211
## 57 2.009748 0.0444578278 0.329977638
## 62 2.204294 0.0275036550 0.279250311TA16.prev.r5$summary[which(TA16.prev.r5$summary$covariates != "(Intercept)" & TA16.prev.r5$summary$p.value < 0.05),]## resp_var covariates estimate std.error statistic
## 30 ProtostrongylideLarver01 Exact_age -2.232596 0.5807771 -3.844153
## 37 EalcesLarvae01 Exact_age -3.658237 1.0696693 -3.419970
## 44 ValcesLarvae01 Exact_age -1.314280 0.4489651 -2.927354
## p.value p.adj
## 30 0.0001209695 0.005927505
## 37 0.0006262795 0.015343849
## 44 0.0034185954 0.055837058TA16.prev.r6$summary[which(TA16.prev.r6$summary$covariates != "(Intercept)" & TA16.prev.r6$summary$p.value < 0.05),]## resp_var covariates estimate std.error statistic
## 30 ProtostrongylideLarver01 Exact_age -2.352179 0.7287718 -3.227593
## 37 EalcesLarvae01 Exact_age -3.170190 0.8669529 -3.656704
## 44 ValcesLarvae01 Exact_age -1.444032 0.5841358 -2.472083
## p.value p.adj
## 30 0.0012483643 0.02038995
## 37 0.0002554792 0.01251848
## 44 0.0134328208 0.10970137# posthoc pairwise comparisons for variables significant for Municipality
TA16.prev.r4.posthoc.Trichuris <- glht(TA16.prev.r4$models$Trichuris01, linfct = mcp(Municipality = "Tukey"))
TA16.prev.r4.posthoc.Valces <- glht(TA16.prev.r4$models$ValcesLarvae01, linfct = mcp(Municipality = "Tukey"))plot(allEffects(TA16.prev.r1$models$ProtostrongylideLarver01))
plot(allEffects(TA16.prev.r1$models$EalcesLarvae01))
plot(allEffects(TA16.prev.r1$models$ValcesLarvae01))
plot(allEffects(TA16.prev.r2$models$Trichuris01))
plot(allEffects(TA16.prev.r2$models$ProtostrongylideLarver01))
plot(allEffects(TA16.prev.r2$models$EalcesLarvae01))
plot(allEffects(TA16.prev.r2$models$ValcesLarvae01))
plot(allEffects(TA16.prev.r3$models$EalcesLarvae01))
plot(allEffects(TA16.prev.r4$models$EalcesLarvae01))
plot(allEffects(TA16.prev.r6$models$ProtostrongylideLarver01))
plot(allEffects(TA16.prev.r6$models$EalcesLarvae01))
plot(allEffects(TA16.prev.r6$models$ValcesLarvae01))
Model selection for key parasite groups
TA16.prev.aic <- rbind.data.frame(
data.frame(resp_var="ProtostrongylideLarver01",
AIC(TA16.prev.r1$models$ProtostrongylideLarver01,
TA16.prev.r2$models$ProtostrongylideLarver01,
TA16.prev.r3$models$ProtostrongylideLarver01,
TA16.prev.r4$models$ProtostrongylideLarver01,
TA16.prev.r5$models$ProtostrongylideLarver01,
TA16.prev.r6$models$ProtostrongylideLarver01)),
data.frame(resp_var="EalcesLarvae01",
AIC(TA16.prev.r1$models$EalcesLarvae01,
TA16.prev.r2$models$EalcesLarvae01,
TA16.prev.r3$models$EalcesLarvae01,
TA16.prev.r4$models$EalcesLarvae01,
TA16.prev.r5$models$EalcesLarvae01,
TA16.prev.r6$models$EalcesLarvae01)),
data.frame(resp_var="ValcesLarvae01",
AIC(TA16.prev.r1$models$ValcesLarvae01,
TA16.prev.r2$models$ValcesLarvae01,
TA16.prev.r3$models$ValcesLarvae01,
TA16.prev.r4$models$ValcesLarvae01,
TA16.prev.r5$models$ValcesLarvae01,
TA16.prev.r6$models$ValcesLarvae01))
)
TA16.prev.aic[order(TA16.prev.aic$resp_var, TA16.prev.aic$AIC),]## resp_var df
## TA16.prev.r2$models$EalcesLarvae01 EalcesLarvae01 6
## TA16.prev.r3$models$EalcesLarvae01 EalcesLarvae01 6
## TA16.prev.r5$models$EalcesLarvae01 EalcesLarvae01 7
## TA16.prev.r6$models$EalcesLarvae01 EalcesLarvae01 7
## TA16.prev.r4$models$EalcesLarvae01 EalcesLarvae01 9
## TA16.prev.r1$models$EalcesLarvae01 EalcesLarvae01 6
## TA16.prev.r2$models$ProtostrongylideLarver01 ProtostrongylideLarver01 6
## TA16.prev.r3$models$ProtostrongylideLarver01 ProtostrongylideLarver01 6
## TA16.prev.r5$models$ProtostrongylideLarver01 ProtostrongylideLarver01 7
## TA16.prev.r1$models$ProtostrongylideLarver01 ProtostrongylideLarver01 6
## TA16.prev.r6$models$ProtostrongylideLarver01 ProtostrongylideLarver01 7
## TA16.prev.r4$models$ProtostrongylideLarver01 ProtostrongylideLarver01 9
## TA16.prev.r2$models$ValcesLarvae01 ValcesLarvae01 6
## TA16.prev.r3$models$ValcesLarvae01 ValcesLarvae01 6
## TA16.prev.r4$models$ValcesLarvae01 ValcesLarvae01 9
## TA16.prev.r6$models$ValcesLarvae01 ValcesLarvae01 7
## TA16.prev.r5$models$ValcesLarvae01 ValcesLarvae01 7
## TA16.prev.r1$models$ValcesLarvae01 ValcesLarvae01 6
## AIC
## TA16.prev.r2$models$EalcesLarvae01 86.28170
## TA16.prev.r3$models$EalcesLarvae01 86.28615
## TA16.prev.r5$models$EalcesLarvae01 86.40806
## TA16.prev.r6$models$EalcesLarvae01 88.02912
## TA16.prev.r4$models$EalcesLarvae01 88.69450
## TA16.prev.r1$models$EalcesLarvae01 91.32413
## TA16.prev.r2$models$ProtostrongylideLarver01 88.34388
## TA16.prev.r3$models$ProtostrongylideLarver01 88.69656
## TA16.prev.r5$models$ProtostrongylideLarver01 88.94797
## TA16.prev.r1$models$ProtostrongylideLarver01 89.65814
## TA16.prev.r6$models$ProtostrongylideLarver01 90.19641
## TA16.prev.r4$models$ProtostrongylideLarver01 92.52291
## TA16.prev.r2$models$ValcesLarvae01 113.71232
## TA16.prev.r3$models$ValcesLarvae01 114.10209
## TA16.prev.r4$models$ValcesLarvae01 115.23682
## TA16.prev.r6$models$ValcesLarvae01 115.54372
## TA16.prev.r5$models$ValcesLarvae01 115.71126
## TA16.prev.r1$models$ValcesLarvae01 117.81090# EalcesLarvae01: r2/r3
# ProtostrongylideLarver01: r2c, then r3
# ValcesLarvae01: r2, then r3Add no. of positive samples to models
TA16.prev.r1$summary$n.pos <- sapply(TA16.prev.r1$summary$resp_var, function(x){
sum(TA16.sub[,x], na.rm = T)
})
TA16.prev.r2$summary$n.pos <- sapply(TA16.prev.r2$summary$resp_var, function(x){
sum(TA16.sub[,x], na.rm = T)
})
TA16.prev.r3$summary$n.pos <- sapply(TA16.prev.r3$summary$resp_var, function(x){
sum(TA16.sub[,x], na.rm = T)
})
TA16.prev.r4$summary$n.pos <- sapply(TA16.prev.r4$summary$resp_var, function(x){
sum(TA16.sub[,x], na.rm = T)
})
TA16.prev.r5$summary$n.pos <- sapply(TA16.prev.r5$summary$resp_var, function(x){
sum(TA16.sub[,x], na.rm = T)
})
TA16.prev.r6$summary$n.pos <- sapply(TA16.prev.r6$summary$resp_var, function(x){
sum(TA16.sub[,x], na.rm = T)
})TA16.prev.r2.resplotA <-
ggplot(
data.frame(
fitted=TA16.prev.r2$models$ProtostrongylideLarver01$fitted.values,
residual=TA16.prev.r2$models$ProtostrongylideLarver01$residuals
),
aes(fitted, residual))+
geom_hline(yintercept = 0, colour = "grey50", linewidth = 0.5, linetype = 2)+
geom_point()+
geom_smooth(method = "loess")+
labs(x = "Fitted values", y = "Residuals", title = "Protostrongylidae detection")+
theme_classic()+
theme(plot.title = element_text(face = "bold", hjust = 0.5))
TA16.prev.r2.resplotB <-
ggplot(
data.frame(
fitted=TA16.prev.r2$models$EalcesLarvae01$fitted.values,
residual=TA16.prev.r2$models$EalcesLarvae01$residuals
),
aes(fitted, residual))+
geom_hline(yintercept = 0, colour = "grey50", linewidth = 0.5, linetype = 2)+
geom_point()+
geom_smooth(method = "loess")+
labs(x = "Fitted values", y = "Residuals", title = "E. alces detection")+
theme_classic()+
theme(plot.title = element_text(face = "bold", hjust = 0.5))
TA16.prev.r2.resplotC <-
ggplot(
data.frame(
fitted=TA16.prev.r2$models$ValcesLarvae01$fitted.values,
residual=TA16.prev.r2$models$ValcesLarvae01$residuals
),
aes(fitted, residual))+
geom_hline(yintercept = 0, colour = "grey50", linewidth = 0.5, linetype = 2)+
geom_point()+
geom_smooth(method = "loess")+
labs(x = "Fitted values", y = "Residuals", title = "V. alces detection")+
theme_classic()+
theme(plot.title = element_text(face = "bold", hjust = 0.5))plot_grid(TA16.prev.r2.resplotA, TA16.prev.r2.resplotB, TA16.prev.r2.resplotC,
nrow = 1)## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
ggplot(subset(TA16.det.df, variable %in% c("ProtostrongylideLarver01","EalcesLarvae01","ValcesLarvae01")),
aes(Adult_age_category, value))+
geom_violin(aes(fill = Adult_age_category), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2, jitter.height = 0.05))+
scale_fill_manual(values = c("#C7E9C0","#74C476","#006D2C"))+
scale_x_discrete(labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
labs(x = "Age group", y = "Detection", fill = "", title = "Trøndelag autumn 2016")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "none",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable, nrow = 1)
# E & V alces only really relevant for agegroup=1
ggplot(subset(TA16.det.df, variable %in% c("EalcesLarvae01","ValcesLarvae01")),
aes(Sex, value))+
geom_violin(aes(fill = Carcass_mass_category), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2, jitter.height = 0.05))+
scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
# scale_x_discrete(labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
labs(x = "Sex", y = "Detection", fill = "Weight category", title = "Trøndelag høst 2016")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable + Adult_age_category, nrow = 1)
ggplot(subset(TA16.det.df, variable %in% c("ValcesLarvae01") & Adult_age_category == "agegroup1"),
aes(Municipality, value))+
geom_violin(aes(fill = Municipality), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2, jitter.height = 0.05))+
geom_signif(comparisons = list(c("Malvik","Tydal")), annotations = c("."),
tip_length = 0.05, y_position = 1.1)+
scale_y_continuous(expand = expansion(mult = c(0, 0.15)))+
labs(x = "Municipality", y = "Detection", fill = "", title = "Trøndelag høst 2016")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "none",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 90))+
facet_wrap(~ variable + Adult_age_category, nrow = 3)
Fig. 2
Point plot with 95% CI:
TA16.det.by.age <- dcast(
TA16.det.df,
Adult_age_category ~ variable,
value.var = "value",
fun.aggregate = sum, fill = 0, drop = T, na.rm = T
)
## Calculate no. of individuals based on samples for which test was performed
TA16.det.by.age$num_inds.MM <- sapply(seq(1,nrow(TA16.det.by.age)), function(i){
age <- TA16.det.by.age[i,"Adult_age_category"]
length(which(TA16.sub$Adult_age_category == age & TA16.sub$McMaster_performed == 1))
})
TA16.det.by.age$num_inds.BA <- sapply(seq(1,nrow(TA16.det.by.age)), function(i){
age <- TA16.det.by.age[i,"Adult_age_category"]
length(which(TA16.sub$Adult_age_category == age & TA16.sub$Baermann_performed == 1))
})
TA16.det.by.age.hm <- melt(TA16.det.by.age,
id.vars = c("Adult_age_category","num_inds.MM","num_inds.BA"),
measure.vars = c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
variable.name = "parasite", value.name = "n.positive")
TA16.det.by.age.hm$proportion <- ifelse(
TA16.det.by.age.hm$parasite %in% mcmaster.parasite.groups.01,
TA16.det.by.age.hm$n.positive / TA16.det.by.age.hm$num_inds.MM,
TA16.det.by.age.hm$n.positive / TA16.det.by.age.hm$num_inds.BA
)
# add 95%CI to prevalence
TA16.det.by.age.hm$CI95.lower <- sapply(seq(1,nrow(TA16.det.by.age.hm)), function(i){
p <- as.character(TA16.det.by.age.hm[i,"parasite"])
n.det <- TA16.det.by.age.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TA16.det.by.age.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TA16.det.by.age.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[1])
})
TA16.det.by.age.hm$CI95.upper <- sapply(seq(1,nrow(TA16.det.by.age.hm)), function(i){
p <- as.character(TA16.det.by.age.hm[i,"parasite"])
n.det <- TA16.det.by.age.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TA16.det.by.age.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TA16.det.by.age.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[2])
})
TA16.det.by.age.hm$CI95.lower.v2 <- ifelse(TA16.det.by.age.hm$proportion == 0,NA,TA16.det.by.age.hm$CI95.lower)
TA16.det.by.age.hm$CI95.upper.v2 <- ifelse(TA16.det.by.age.hm$proportion == 0,NA,TA16.det.by.age.hm$CI95.upper)
TA16.det.by.age.hm$variable <- factor(TA16.det.by.age.hm$parasite,
levels = levels(TA16.det.df$variable))TA16.plot.detA <-
ggplot(
subset(TA16.det.by.age.hm, variable %in% c("ProtostrongylideLarver01","EalcesLarvae01","ValcesLarvae01")))+
geom_errorbar(aes(x = Adult_age_category, ymin = CI95.lower.v2, ymax = CI95.upper.v2),
width = 0.5, linewidth = 0.75, colour = "black")+
geom_point(aes(Adult_age_category, proportion, fill = Adult_age_category, size = num_inds.BA), shape = 21)+
scale_fill_manual(values = c("#C7E9C0","#74C476","#006D2C"), labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"), guide = "none")+
scale_x_discrete(labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
labs(x = "Age group", y = "Prevalence", size = "Total no.\nof samples")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = ggtext::element_markdown(size = 13),
legend.position = "right",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(ProtostrongylideLarver01="Protostrongylidae", EalcesLarvae01="*E. alces*", ValcesLarvae01="*V. alces*")))TA16.det.by.KgCat <- dcast(
TA16.det.df,
Adult_age_category + Carcass_mass_category ~ variable,
value.var = "value",
fun.aggregate = sum, fill = 0, drop = T, na.rm = T
)
## Calculate no. of individuals based on samples for which test was performed
TA16.det.by.KgCat$num_inds.MM <- sapply(seq(1,nrow(TA16.det.by.KgCat)), function(i){
age <- TA16.det.by.KgCat[i,"Adult_age_category"]
kgC <- TA16.det.by.KgCat[i,"Carcass_mass_category"]
length(which(TA16.sub$Adult_age_category == age & TA16.sub$Carcass_mass_category == kgC & TA16.sub$McMaster_performed == 1))
})
TA16.det.by.KgCat$num_inds.BA <- sapply(seq(1,nrow(TA16.det.by.KgCat)), function(i){
age <- TA16.det.by.KgCat[i,"Adult_age_category"]
kgC <- TA16.det.by.KgCat[i,"Carcass_mass_category"]
length(which(TA16.sub$Adult_age_category == age & TA16.sub$Carcass_mass_category == kgC & TA16.sub$Baermann_performed == 1))
})
TA16.det.by.KgCat.hm <- melt(TA16.det.by.KgCat,
id.vars = c("Adult_age_category","Carcass_mass_category","num_inds.MM","num_inds.BA"),
measure.vars = c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
variable.name = "parasite", value.name = "n.positive")
TA16.det.by.KgCat.hm$proportion <- ifelse(
TA16.det.by.KgCat.hm$parasite %in% mcmaster.parasite.groups.01,
TA16.det.by.KgCat.hm$n.positive / TA16.det.by.KgCat.hm$num_inds.MM,
TA16.det.by.KgCat.hm$n.positive / TA16.det.by.KgCat.hm$num_inds.BA
)
# add 95%CI to prevalence
TA16.det.by.KgCat.hm$CI95.lower <- sapply(seq(1,nrow(TA16.det.by.KgCat.hm)), function(i){
p <- as.character(TA16.det.by.KgCat.hm[i,"parasite"])
n.det <- TA16.det.by.KgCat.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TA16.det.by.KgCat.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TA16.det.by.KgCat.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[1])
})
TA16.det.by.KgCat.hm$CI95.upper <- sapply(seq(1,nrow(TA16.det.by.KgCat.hm)), function(i){
p <- as.character(TA16.det.by.KgCat.hm[i,"parasite"])
n.det <- TA16.det.by.KgCat.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TA16.det.by.KgCat.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TA16.det.by.KgCat.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[2])
})
TA16.det.by.KgCat.hm$CI95.lower.v2 <- ifelse(TA16.det.by.KgCat.hm$proportion == 0,NA,TA16.det.by.KgCat.hm$CI95.lower)
TA16.det.by.KgCat.hm$CI95.upper.v2 <- ifelse(TA16.det.by.KgCat.hm$proportion == 0,NA,TA16.det.by.KgCat.hm$CI95.upper)
TA16.det.by.KgCat.hm$variable <- factor(TA16.det.by.KgCat.hm$parasite,
levels = levels(TA16.det.df$variable))
# TA16.det.by.KgCat.hm$group <- factor(paste(TA16.det.by.KgCat.hm$Carcass_mass_category, TA16.det.by.KgCat.hm$Sex, sep = "_"),
# levels = c("Normal_Male","Low_Male","Normal_Female","Low_Female"))TA16.plot.detB <-
ggplot(
subset(TA16.det.by.KgCat.hm, variable %in% c("ProtostrongylideLarver01","EalcesLarvae01","ValcesLarvae01") & Adult_age_category == "agegroup1"))+
geom_errorbar(aes(x = Carcass_mass_category, ymin = CI95.lower.v2, ymax = CI95.upper.v2),
width = 0.5, linewidth = 0.75, colour = "black")+
geom_point(aes(Carcass_mass_category, proportion, fill = Carcass_mass_category, size = num_inds.BA), shape = 21)+
scale_fill_manual(values = c("#3C2692","#CBC9DA"), guide = "none")+
scale_size(limits = c(10,60), range = c(1,5))+
labs(x = "Carcass mass category", y = "Prevalence", size = "Total no.\nof samples", fill = "Carcass mass\ncategory")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = ggtext::element_markdown(size = 13),
legend.position = "right", legend.box = "horizontal",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable + Adult_age_category, nrow = 1,
labeller = labeller(variable = c(ProtostrongylideLarver01="Protostrongylidae", EalcesLarvae01="*E. alces*", ValcesLarvae01="*V. alces*"),
Adult_age_category = c(agegroup1 = "1 yr")))plot_grid(TA16.plot.detA, TA16.plot.detB,
align = "hv", axis = "tblr", nrow = 2,
labels = c("a","b"))
ggsave("figures/Fig2_PrevalenceDSL_by_AgeMassCat.png",units = "in", dpi = 1200, width = 9, height = 6.5)Comparison of carcass mass (ns) vs carcass mass category (sig) for EalcesLarvae01 & ValcesLarvae01
ggplot(
data.frame(
variable=c("EalcesLarvae01","EalcesLarvae01",
"ValcesLarvae01","ValcesLarvae01"),
group=c("Normal","Low","Normal","Low"),
proportion=c(38/(38+15),13/(13+1),
33/(33+20),11/(11+3)),
CI95.lower=c(binom.test(38, (38+15), conf.level = 0.95)$conf.int[1],
binom.test(13, (13+1), conf.level = 0.95)$conf.int[1],
binom.test(33, (33+20), conf.level = 0.95)$conf.int[1],
binom.test(11, (11+3), conf.level = 0.95)$conf.int[1]),
CI95.upper=c(binom.test(38, (38+15), conf.level = 0.95)$conf.int[2],
binom.test(13, (13+1), conf.level = 0.95)$conf.int[2],
binom.test(33, (33+20), conf.level = 0.95)$conf.int[2],
binom.test(11, (11+3), conf.level = 0.95)$conf.int[2])
))+
geom_errorbar(aes(x = group, ymin = CI95.lower, ymax = CI95.upper),
width = 0.5, linewidth = 0.75, colour = "black")+
geom_point(aes(group, proportion, fill = group), shape = 21, size = 4)+
scale_fill_manual(values = c("#CBC9DA", "#3C2692"))+
labs(x = "Carcass mass category", y = "Prevalence", size = "Total no.\nof samples", fill = "Carcass mass\ncategory", title = "Yearlings")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = ggtext::element_markdown(size = 13),
legend.position = "none", legend.box = "horizontal",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(EalcesLarvae01="*E. alces*", ValcesLarvae01="*V. alces*")))
ggplot(subset(TA16.det.df, variable %in% c("EalcesLarvae01","ValcesLarvae01") & Adult_age_category == "agegroup1" & !is.na(value)),
aes(as.factor(value), Carcass_mass_kg))+
geom_hline(yintercept = 106, linewidth = 1, colour = "grey50", linetype = 2)+
geom_boxplot(aes(fill = as.factor(value)), alpha = 0.25, outlier.colour = NA)+
geom_point(aes(fill = as.factor(value)), shape = 21, color = "black", size = 3)+
scale_x_discrete(labels = c("No","Yes"))+
scale_fill_manual(values = c("grey25","grey75"))+
labs(x = "Parasite detected", y = "Carcass mass (kg)", title = "Yearlings")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(EalcesLarvae01="E. alces", ValcesLarvae01="V. alces")))
# ggplot(subset(TA16.det.df, variable %in% c("EalcesLarvae01","ValcesLarvae01") & Adult_age_category == "agegroup1"),
# aes(Carcass_mass_kg, value))+
# geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 3)+
# scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
# labs(x = "Carcass mass (kg)", y = "Detected (binary)", title = "Yearlings")+
# theme_classic()+
# theme(plot.title = element_text(size = 13, hjust = 0.5),
# axis.text = element_text(color = "black", size = 12),
# axis.title = element_text(size = 13),
# axis.text.x = element_text(angle = 0),
# strip.text.x = element_text(size = 13),
# legend.position = "none",
# legend.text = element_text(color = "black", size = 12),
# legend.title = element_text(size = 13))+
# facet_wrap(~ variable, nrow = 2,
# labeller = labeller(variable = c(EalcesLarvae01="E. alces", ValcesLarvae01="V. alces")))Abundance
Must be present in at least 5 samples to test (otherwise not enough variation in dataset). Rounding abundances to nearest integer and running negative binomial GLMs with nbinom1. Six models per parasite:
- exact age + sex + weight + municipality
- age group + sex + weight category + municipality
- age group + sex * weight category + municipality
- exact age * weight category + sex + municipality
- age group + sex + weight category + harvest density (2016)
- age group + sex + weight category + harvest density (mean over 2016-2018)
TA16.abund.r2 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.EPG,baermann.parasite.groups.LPG),
" ~ Exact_age + Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "nbinom1",
ct_round = T,
min.samples = 5
)## [1] "Skipping EimeriaOPG, not enough positive samples."
## [1] "Running Neg Binomial model for StrongylidetypeeggEPG"
## [1] "Running Neg Binomial model for NematodirussppEPG"
## [1] "Skipping NematodirusBattusEPG, not enough positive samples."
## [1] "Running Neg Binomial model for TrichurisEPG"
## [1] "Skipping CapillariaEPG, not enough positive samples."
## [1] "Skipping MiddEggEPG, not enough positive samples."
## [1] "Running Neg Binomial model for ProtostrongylideLarverLPG"
## [1] "Skipping DictyocaulusLPG, not enough positive samples."# kg continuous
TA16.abund.r1 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.EPG,baermann.parasite.groups.LPG),
" ~ Exact_age + Carcass_mass_kg + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "nbinom1",
ct_round = T,
min.samples = 5
)## [1] "Skipping EimeriaOPG, not enough positive samples."
## [1] "Running Neg Binomial model for StrongylidetypeeggEPG"
## [1] "Running Neg Binomial model for NematodirussppEPG"
## [1] "Skipping NematodirusBattusEPG, not enough positive samples."
## [1] "Running Neg Binomial model for TrichurisEPG"
## [1] "Skipping CapillariaEPG, not enough positive samples."
## [1] "Skipping MiddEggEPG, not enough positive samples."
## [1] "Running Neg Binomial model for ProtostrongylideLarverLPG"
## [1] "Skipping DictyocaulusLPG, not enough positive samples."# age group
# TA16.abund.r2 <- run_glm(
# TA16.sub,
# c(mcmaster.parasite.groups.EPG,baermann.parasite.groups.LPG),
# " ~ Adult_age_category + Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18",
# fam = "nbinom1",
# ct_round = T,
# min.samples = 5
# )
# 2016 density
TA16.abund.r3 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.EPG,baermann.parasite.groups.LPG),
" ~ Exact_age + Carcass_mass_category + Sex + MunicipalityDensity_2016 + Days.harvested.vs.lab",
fam = "nbinom1",
ct_round = T,
min.samples = 5
)## [1] "Skipping EimeriaOPG, not enough positive samples."
## [1] "Running Neg Binomial model for StrongylidetypeeggEPG"
## [1] "Running Neg Binomial model for NematodirussppEPG"
## [1] "Skipping NematodirusBattusEPG, not enough positive samples."
## [1] "Running Neg Binomial model for TrichurisEPG"
## [1] "Skipping CapillariaEPG, not enough positive samples."
## [1] "Skipping MiddEggEPG, not enough positive samples."
## [1] "Running Neg Binomial model for ProtostrongylideLarverLPG"
## [1] "Skipping DictyocaulusLPG, not enough positive samples."# Municipality
TA16.abund.r4 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.EPG,baermann.parasite.groups.LPG),
" ~ Exact_age + Carcass_mass_category + Sex + Municipality + Days.harvested.vs.lab",
fam = "nbinom1",
ct_round = T,
min.samples = 5
)## [1] "Skipping EimeriaOPG, not enough positive samples."
## [1] "Running Neg Binomial model for StrongylidetypeeggEPG"
## [1] "Running Neg Binomial model for NematodirussppEPG"
## [1] "Skipping NematodirusBattusEPG, not enough positive samples."
## [1] "Running Neg Binomial model for TrichurisEPG"
## [1] "Skipping CapillariaEPG, not enough positive samples."
## [1] "Skipping MiddEggEPG, not enough positive samples."
## [1] "Running Neg Binomial model for ProtostrongylideLarverLPG"
## [1] "Skipping DictyocaulusLPG, not enough positive samples."# sex x weight interaction term
TA16.abund.r5 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.EPG,baermann.parasite.groups.LPG),
" ~ Exact_age + Carcass_mass_category * Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "nbinom1",
ct_round = T,
min.samples = 5
)## [1] "Skipping EimeriaOPG, not enough positive samples."
## [1] "Running Neg Binomial model for StrongylidetypeeggEPG"
## [1] "Running Neg Binomial model for NematodirussppEPG"
## [1] "Skipping NematodirusBattusEPG, not enough positive samples."
## [1] "Running Neg Binomial model for TrichurisEPG"
## [1] "Skipping CapillariaEPG, not enough positive samples."
## [1] "Skipping MiddEggEPG, not enough positive samples."
## [1] "Running Neg Binomial model for ProtostrongylideLarverLPG"
## [1] "Skipping DictyocaulusLPG, not enough positive samples."# age x kg cat interaction term
TA16.abund.r6 <- run_glm(
TA16.sub,
c(mcmaster.parasite.groups.EPG,baermann.parasite.groups.LPG),
" ~ Exact_age * Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "nbinom1",
ct_round = T,
min.samples = 5
)## [1] "Skipping EimeriaOPG, not enough positive samples."
## [1] "Running Neg Binomial model for StrongylidetypeeggEPG"
## [1] "Running Neg Binomial model for NematodirussppEPG"
## [1] "Skipping NematodirusBattusEPG, not enough positive samples."
## [1] "Running Neg Binomial model for TrichurisEPG"
## [1] "Skipping CapillariaEPG, not enough positive samples."
## [1] "Skipping MiddEggEPG, not enough positive samples."
## [1] "Running Neg Binomial model for ProtostrongylideLarverLPG"
## [1] "Skipping DictyocaulusLPG, not enough positive samples."TA16.abund.r1$summary[which(TA16.abund.r1$summary$covariates != "(Intercept)" & TA16.abund.r1$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 20 ProtostrongylideLarverLPG Exact_age -1.24104385 0.397741012
## 21 ProtostrongylideLarverLPG Carcass_mass_kg -0.01804358 0.006941988
## 24 ProtostrongylideLarverLPG Days.harvested.vs.lab 0.08905558 0.025821339
## statistic p.value p.adj
## 20 -3.120231 0.0018070925 0.010842555
## 21 -2.599194 0.0093442840 0.044852563
## 24 3.448914 0.0005628456 0.004502765TA16.abund.r2$summary[which(TA16.abund.r2$summary$covariates != "(Intercept)" & TA16.abund.r2$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 3 StrongylidetypeeggEPG Carcass_mass_category.L 0.24953999 0.12403997
## 4 StrongylidetypeeggEPG SexFemale -0.33399104 0.16034083
## 20 ProtostrongylideLarverLPG Exact_age -1.65800204 0.38520675
## 21 ProtostrongylideLarverLPG Carcass_mass_category.L 0.74932746 0.17768185
## 22 ProtostrongylideLarverLPG SexFemale -0.54856357 0.24023279
## 24 ProtostrongylideLarverLPG Days.harvested.vs.lab 0.09611575 0.02484926
## statistic p.value p.adj
## 3 2.011771 4.424411e-02 0.1327323186
## 4 -2.083007 3.725061e-02 0.1277163640
## 20 -4.304187 1.675997e-05 0.0001340798
## 21 4.217243 2.473079e-05 0.0001483847
## 22 -2.283467 2.240290e-02 0.0896115807
## 24 3.867952 1.097532e-04 0.0005268155TA16.abund.r3$summary[which(TA16.abund.r3$summary$covariates != "(Intercept)" & TA16.abund.r3$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 4 StrongylidetypeeggEPG SexFemale -0.32534254 0.16032298
## 20 ProtostrongylideLarverLPG Exact_age -1.68398890 0.38703777
## 21 ProtostrongylideLarverLPG Carcass_mass_category.L 0.73711894 0.17553713
## 22 ProtostrongylideLarverLPG SexFemale -0.55551365 0.23927322
## 24 ProtostrongylideLarverLPG Days.harvested.vs.lab 0.09512646 0.02471322
## statistic p.value p.adj
## 4 -2.029294 4.242830e-02 0.1454684717
## 20 -4.350968 1.355379e-05 0.0001084303
## 21 4.199220 2.678365e-05 0.0001607019
## 22 -2.321671 2.025067e-02 0.0810026748
## 24 3.849213 1.184978e-04 0.0005687894TA16.abund.r4$summary[which(TA16.abund.r4$summary$covariates != "(Intercept)" & TA16.abund.r4$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 3 StrongylidetypeeggEPG Carcass_mass_category.L 0.23694734 0.12070813
## 4 StrongylidetypeeggEPG SexFemale -0.35246577 0.16050113
## 6 StrongylidetypeeggEPG MunicipalitySelbu 0.90794612 0.33869841
## 24 TrichurisEPG MunicipalitySelbu -1.63560186 0.68257673
## 29 ProtostrongylideLarverLPG Exact_age -1.91127716 0.39672676
## 30 ProtostrongylideLarverLPG Carcass_mass_category.L 0.79327464 0.17718616
## 31 ProtostrongylideLarverLPG SexFemale -0.57447542 0.24157287
## 34 ProtostrongylideLarverLPG MunicipalityStjørdal 1.55677808 0.69726554
## 36 ProtostrongylideLarverLPG Days.harvested.vs.lab 0.09055432 0.02434742
## statistic p.value p.adj
## 3 1.962977 4.964879e-02 1.378070e-01
## 4 -2.196033 2.808958e-02 9.192955e-02
## 6 2.680692 7.347011e-03 3.778463e-02
## 24 -2.396217 1.656529e-02 6.961547e-02
## 29 -4.817616 1.452837e-06 1.743404e-05
## 30 4.477069 7.567481e-06 6.810733e-05
## 31 -2.378063 1.740387e-02 6.961547e-02
## 34 2.232690 2.556937e-02 9.192955e-02
## 36 3.719257 1.998095e-04 1.438628e-03TA16.abund.r5$summary[which(TA16.abund.r5$summary$covariates != "(Intercept)" & TA16.abund.r5$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 23 ProtostrongylideLarverLPG Exact_age -1.6591378 0.3860986
## 24 ProtostrongylideLarverLPG Carcass_mass_category.L 0.7871043 0.2260995
## 25 ProtostrongylideLarverLPG SexFemale -0.5617693 0.2438951
## 27 ProtostrongylideLarverLPG Days.harvested.vs.lab 0.0957379 0.0249543
## statistic p.value p.adj
## 23 -4.297186 1.729797e-05 0.0001614478
## 24 3.481229 4.991183e-04 0.0027950625
## 25 -2.303323 2.126067e-02 0.0992164731
## 27 3.836530 1.247852e-04 0.0008734961TA16.abund.r6$summary[which(TA16.abund.r6$summary$covariates != "(Intercept)" & TA16.abund.r6$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 4 StrongylidetypeeggEPG SexFemale -0.33472266 0.16033347
## 23 ProtostrongylideLarverLPG Exact_age -1.75138568 0.47068866
## 25 ProtostrongylideLarverLPG SexFemale -0.55238972 0.23982323
## 27 ProtostrongylideLarverLPG Days.harvested.vs.lab 0.09633104 0.02483006
## statistic p.value p.adj
## 4 -2.087666 0.0368280105 0.1718640488
## 23 -3.720901 0.0001985136 0.0013895953
## 25 -2.303320 0.0212608256 0.1190606233
## 27 3.879614 0.0001046222 0.0009764742TA16.abund.r4.posthoc.Strongylide <- glht(TA16.abund.r4$models$StrongylidetypeeggEPG, linfct = mcp(Municipality = "Tukey"))
summary(TA16.abund.r4.posthoc.Strongylide)##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: glmmTMB(formula = y_rounded ~ Exact_age + Carcass_mass_category +
## Sex + Municipality + Days.harvested.vs.lab, data = df, family = nbinom1,
## na.action = na.omit, ziformula = ~0, dispformula = ~1)
##
## Linear Hypotheses:
## Estimate Std. Error z value Pr(>|z|)
## Meråker - Malvik == 0 0.72487 0.44306 1.636 0.4523
## Selbu - Malvik == 0 0.90795 0.33870 2.681 0.0523 .
## Stjørdal - Malvik == 0 0.66273 0.44284 1.497 0.5427
## Tydal - Malvik == 0 0.78425 0.40632 1.930 0.2831
## Selbu - Meråker == 0 0.18308 0.30578 0.599 0.9726
## Stjørdal - Meråker == 0 -0.06214 0.41428 -0.150 0.9999
## Tydal - Meråker == 0 0.05939 0.37042 0.160 0.9998
## Stjørdal - Selbu == 0 -0.24522 0.30307 -0.809 0.9209
## Tydal - Selbu == 0 -0.12369 0.24571 -0.503 0.9856
## Tydal - Stjørdal == 0 0.12152 0.36943 0.329 0.9972
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)TA16.abund.r4.posthoc.Strongylide.summary <- format_posthoc_summary(TA16.abund.r4.posthoc.Strongylide, "StrongylidetypeeggEPG")
TA16.abund.r4.posthoc.Trichuris <- glht(TA16.abund.r4$models$TrichurisEPG, linfct = mcp(Municipality = "Tukey"))
summary(TA16.abund.r4.posthoc.Trichuris)##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: glmmTMB(formula = y_rounded ~ Exact_age + Carcass_mass_category +
## Sex + Municipality + Days.harvested.vs.lab, data = df, family = nbinom1,
## na.action = na.omit, ziformula = ~0, dispformula = ~1)
##
## Linear Hypotheses:
## Estimate Std. Error z value Pr(>|z|)
## Meråker - Malvik == 0 -1.65521 1.17814 -1.405 0.607
## Selbu - Malvik == 0 -1.63560 0.68258 -2.396 0.109
## Stjørdal - Malvik == 0 -1.10809 0.94273 -1.175 0.752
## Tydal - Malvik == 0 -1.20587 0.91630 -1.316 0.665
## Selbu - Meråker == 0 0.01961 1.05800 0.019 1.000
## Stjørdal - Meråker == 0 0.54713 1.25521 0.436 0.992
## Tydal - Meråker == 0 0.44934 1.19764 0.375 0.995
## Stjørdal - Selbu == 0 0.52752 0.78916 0.668 0.960
## Tydal - Selbu == 0 0.42973 0.69985 0.614 0.971
## Tydal - Stjørdal == 0 -0.09778 0.93821 -0.104 1.000
## (Adjusted p values reported -- single-step method)TA16.abund.r4.posthoc.Protostrongylide <- glht(TA16.abund.r4$models$ProtostrongylideLarverLPG, linfct = mcp(Municipality = "Tukey"))
summary(TA16.abund.r4.posthoc.Protostrongylide)##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: glmmTMB(formula = y_rounded ~ Exact_age + Carcass_mass_category +
## Sex + Municipality + Days.harvested.vs.lab, data = df, family = nbinom1,
## na.action = na.omit, ziformula = ~0, dispformula = ~1)
##
## Linear Hypotheses:
## Estimate Std. Error z value Pr(>|z|)
## Meråker - Malvik == 0 0.57039 0.77240 0.738 0.9416
## Selbu - Malvik == 0 0.50555 0.61889 0.817 0.9175
## Stjørdal - Malvik == 0 1.55678 0.69727 2.233 0.1537
## Tydal - Malvik == 0 0.40860 0.72957 0.560 0.9783
## Selbu - Meråker == 0 -0.06485 0.49745 -0.130 0.9999
## Stjørdal - Meråker == 0 0.98639 0.59456 1.659 0.4349
## Tydal - Meråker == 0 -0.16179 0.61379 -0.264 0.9988
## Stjørdal - Selbu == 0 1.05123 0.35864 2.931 0.0253 *
## Tydal - Selbu == 0 -0.09694 0.42900 -0.226 0.9993
## Tydal - Stjørdal == 0 -1.14818 0.53642 -2.140 0.1869
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)TA16.abund.r4.posthoc.Protostrongylide.summary <- format_posthoc_summary(TA16.abund.r4.posthoc.Protostrongylide, "ProtostrongylideLarverLPG")TA16.abund.aic <- rbind.data.frame(
data.frame(resp_var="ProtostrongylideLarverLPG",
AIC(TA16.abund.r1$models$ProtostrongylideLarverLPG,
TA16.abund.r2$models$ProtostrongylideLarverLPG,
TA16.abund.r3$models$ProtostrongylideLarverLPG,
TA16.abund.r4$models$ProtostrongylideLarverLPG,
TA16.abund.r5$models$ProtostrongylideLarverLPG,
TA16.abund.r6$models$ProtostrongylideLarverLPG)),
data.frame(resp_var="StrongylidetypeeggEPG",
AIC(TA16.abund.r1$models$StrongylidetypeeggEPG,
TA16.abund.r2$models$StrongylidetypeeggEPG,
TA16.abund.r3$models$StrongylidetypeeggEPG,
TA16.abund.r4$models$StrongylidetypeeggEPG,
TA16.abund.r5$models$StrongylidetypeeggEPG,
TA16.abund.r6$models$StrongylidetypeeggEPG))
)
TA16.abund.aic[order(TA16.abund.aic$resp_var, TA16.abund.aic$AIC),]## resp_var df
## TA16.abund.r3$models$ProtostrongylideLarverLPG ProtostrongylideLarverLPG 7
## TA16.abund.r4$models$ProtostrongylideLarverLPG ProtostrongylideLarverLPG 10
## TA16.abund.r2$models$ProtostrongylideLarverLPG ProtostrongylideLarverLPG 7
## TA16.abund.r6$models$ProtostrongylideLarverLPG ProtostrongylideLarverLPG 8
## TA16.abund.r5$models$ProtostrongylideLarverLPG ProtostrongylideLarverLPG 8
## TA16.abund.r1$models$ProtostrongylideLarverLPG ProtostrongylideLarverLPG 7
## TA16.abund.r4$models$StrongylidetypeeggEPG StrongylidetypeeggEPG 10
## TA16.abund.r2$models$StrongylidetypeeggEPG StrongylidetypeeggEPG 7
## TA16.abund.r3$models$StrongylidetypeeggEPG StrongylidetypeeggEPG 7
## TA16.abund.r5$models$StrongylidetypeeggEPG StrongylidetypeeggEPG 8
## TA16.abund.r6$models$StrongylidetypeeggEPG StrongylidetypeeggEPG 8
## TA16.abund.r1$models$StrongylidetypeeggEPG StrongylidetypeeggEPG 7
## AIC
## TA16.abund.r3$models$ProtostrongylideLarverLPG 555.5019
## TA16.abund.r4$models$ProtostrongylideLarverLPG 555.8622
## TA16.abund.r2$models$ProtostrongylideLarverLPG 556.0400
## TA16.abund.r6$models$ProtostrongylideLarverLPG 557.7863
## TA16.abund.r5$models$ProtostrongylideLarverLPG 557.9711
## TA16.abund.r1$models$ProtostrongylideLarverLPG 563.0724
## TA16.abund.r4$models$StrongylidetypeeggEPG 1410.4468
## TA16.abund.r2$models$StrongylidetypeeggEPG 1412.8775
## TA16.abund.r3$models$StrongylidetypeeggEPG 1414.0250
## TA16.abund.r5$models$StrongylidetypeeggEPG 1414.1108
## TA16.abund.r6$models$StrongylidetypeeggEPG 1414.8578
## TA16.abund.r1$models$StrongylidetypeeggEPG 1415.2085# ProtostrongylideLarverLPG: r3/r4/r2
# StrongylidetypeeggEPG: r6/r4/r2cAdd no. of positive samples to selected models
TA16.abund.r1$summary$n.pos <- sapply(TA16.abund.r1$summary$resp_var, function(x){
sum(TA16.sub[,x] > 0, na.rm = T)
})
TA16.abund.r2$summary$n.pos <- sapply(TA16.abund.r2$summary$resp_var, function(x){
sum(TA16.sub[,x] > 0, na.rm = T)
})
TA16.abund.r3$summary$n.pos <- sapply(TA16.abund.r3$summary$resp_var, function(x){
sum(TA16.sub[,x] > 0, na.rm = T)
})
TA16.abund.r4$summary$n.pos <- sapply(TA16.abund.r4$summary$resp_var, function(x){
sum(TA16.sub[,x] > 0, na.rm = T)
})
TA16.abund.r5$summary$n.pos <- sapply(TA16.abund.r5$summary$resp_var, function(x){
sum(TA16.sub[,x] > 0, na.rm = T)
})
TA16.abund.r6$summary$n.pos <- sapply(TA16.abund.r6$summary$resp_var, function(x){
sum(TA16.sub[,x] > 0, na.rm = T)
})TA16.plot.ageA <-
ggplot(subset(TA16.abund.df, variable %in% c("StrongylidetypeeggEPG") &
!is.na(Sex) & !is.na(Adult_age_category) & !is.na(Carcass_mass_category) & !is.na(Municipality)),
aes(Adult_age_category, value))+
geom_boxplot(aes(fill = Adult_age_category), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#C7E9C0","#74C476","#006D2C"))+
scale_x_discrete(labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
scale_y_continuous(transform = "log10")+
labs(x = "Age group", y = "Abundance (log EPG)", fill = "")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(StrongylidetypeeggEPG = "Trichostrongylidae eggs")))
TA16.plot.ageB <-
ggplot(subset(TA16.abund.df, variable %in% c("ProtostrongylideLarverLPG") &
!is.na(Sex) & !is.na(Adult_age_category) & !is.na(Carcass_mass_category) & !is.na(Municipality)),
aes(Adult_age_category, value))+
geom_boxplot(aes(fill = Adult_age_category), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#C7E9C0","#74C476","#006D2C"))+
scale_x_discrete(labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
scale_y_continuous(transform = "log10")+
labs(x = "Age group", y = "Abundance (log LPG)", fill = "")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(ProtostrongylideLarverLPG="Protostrongylidae larvae")))plot_grid(TA16.plot.ageB, TA16.plot.ageA,
nrow = 1, align = "hv", axis = "tblr")
ggplot(subset(TA16.abund.df, variable %in% c("StrongylidetypeeggEPG")),
aes(Municipality, value))+
geom_boxplot(aes(fill = Municipality), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
geom_signif(comparisons = list(c("Malvik","Meråker"),c("Malvik","Selbu"),c("Malvik","Tydal")), annotations = c(".","**","."),
tip_length = 0.04, step_increase = 0.25)+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.15)))+
labs(x = "Municipality", y = "Abundance (log-scale)", fill = "", title = "Trøndelag høst 2016")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "none",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 90))+
facet_wrap(~ variable, nrow = 3)
Fig. 3
TA16.plot.abundA <-
ggplot(subset(TA16.abund.df, variable %in% c("StrongylidetypeeggEPG")),
aes(Carcass_mass_category, value))+
geom_boxplot(aes(fill = Carcass_mass_category), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
labs(x = "Carcass mass category", y = "Abundance (log EPG)", fill = "Carcass mass category")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable + Adult_age_category, nrow = 1,
labeller = labeller(variable = c(StrongylidetypeeggEPG="Trichostrongylidae"),
Adult_age_category = c(agegroup1="1 yr", agegroup2="2-5 yr", agegroup3="6+ yr")))# only really relevant for age group 1
TA16.plot.abundB <-
ggplot(subset(TA16.abund.df, variable %in% c("ProtostrongylideLarverLPG") & Adult_age_category == "agegroup1"),
aes(Carcass_mass_category, value))+
geom_boxplot(aes(fill = Carcass_mass_category), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
labs(x = "Carcass mass category", y = "Abundance (log LPG)", fill = "Carcass mass\ncategory")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable + Adult_age_category, nrow = 1,
labeller = labeller(variable = c(ProtostrongylideLarverLPG="Protostrongylidae"),
Adult_age_category = c(agegroup1="1 yr", agegroup2="2-5 yr", agegroup3="6+ yr")))plot_grid(TA16.plot.abundA, TA16.plot.abundB,
nrow = 1, align = "hv", axis = "tb",
rel_widths = c(1,0.6),
labels = c("a","b"))
ggsave("figures/Fig3_Abundance_by_AgeMassCat.png",units = "in", dpi = 1200, width = 11, height = 4)Fig. S3
Comparison of carcass mass (sig) vs carcass mass category (sig) for ProtostrongylideLarverLPG
ggplot(subset(TA16.abund.df, variable %in% c("ProtostrongylideLarverLPG") & Adult_age_category == "agegroup1"),
aes(Carcass_mass_kg, value))+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 3)+
scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
# scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
scale_y_continuous(limits = c(0,round(max(subset(TA16.abund.df, variable %in% c("ProtostrongylideLarverLPG") & Adult_age_category == "agegroup1")$value, na.rm = T),-1)))+
labs(x = "Carcass mass (kg)", y = "Abundance (LPG)", fill = "Carcass mass\ncategory")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "right",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable + Adult_age_category, nrow = 1,
labeller = labeller(variable = c(ProtostrongylideLarverLPG="Protostrongylidae"),
Adult_age_category = c(agegroup1="1 yr", agegroup2="2-5 yr", agegroup3="6+ yr")))## `geom_smooth()` using formula = 'y ~ x'## Warning: Removed 8 rows containing non-finite outside the scale range
## (`stat_smooth()`).## Warning: Removed 13 rows containing missing values or values outside the scale range
## (`geom_smooth()`).## Warning: Removed 8 rows containing missing values or values outside the scale range
## (`geom_point()`).
ggsave("figures/FigS3_ProtostrongylidaeAbundance_by_MassKg.png",units = "in", dpi = 1200, width = 7.5, height = 4)## `geom_smooth()` using formula = 'y ~ x'## Warning: Removed 8 rows containing non-finite outside the scale range
## (`stat_smooth()`).## Warning: Removed 13 rows containing missing values or values outside the scale range
## (`geom_smooth()`).## Warning: Removed 8 rows containing missing values or values outside the scale range
## (`geom_point()`).ggplot(subset(TA16.abund.df, variable %in% c("TrichurisEPG")),
aes(Carcass_mass_kg, value))+
geom_smooth(method=glm , color="black", fill="grey70", se=TRUE)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3)+
scale_fill_manual(values = c("#C7E9C0","#74C476","#006D2C"), labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
labs(x = "Carcass mass (kg)", y = "Abundance (EPG)", fill = "Age group")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "right",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))
ggplot(subset(TA16.abund.df, variable %in% c("TrichurisEPG")),
aes(Adult_age_category, value))+
# geom_smooth(method=glm , color="black", fill="grey70", se=TRUE)+
# geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 3)+
# scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
# labs(x = "Exact age (years)", y = "Abundance (EPG)", fill = "Carcass mass\ncategory")+
geom_boxplot(aes(fill = Adult_age_category), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#C7E9C0","#74C476","#006D2C"), labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
labs(x = "Age group", y = "Abundance (EPG)", fill = "Age group")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))
Fig. S5
Trichostrongylidae & Protostrongylidae egg abundance and municipality
TA16.sub$Municipality2 <- factor(
TA16.sub$Municipality,
levels = unique(TA16.sub[order(TA16.sub$MunicipalityDensity_Mean2016_18),"Municipality"]))TA16.plot.abundC <-
ggplot(TA16.sub, aes(MunicipalityDensity_Mean2016_18, StrongylidetypeeggEPG))+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
# geom_point(fill = "gold", shape = 21, color = "black", size = 3)+
geom_point(aes(fill = Municipality2), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.height = 0.1, jitter.width = 0.05))+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
scale_x_continuous(limits = c(0.2,0.7))+
labs(x = "Moose density (mean 2016-2018)", y = "Abundance (log EPG)", fill = "")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))
TA16.plot.abundD <-
ggplot(TA16.sub, aes(Municipality2, StrongylidetypeeggEPG))+
geom_violin(aes(fill = Municipality2), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Municipality2), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
geom_signif(comparisons = list(c("Selbu","Malvik")),
annotations = paste0("p = ",
round(TA16.abund.r4.posthoc.Strongylide.summary[which(TA16.abund.r4.posthoc.Strongylide.summary$covariates == "Selbu - Malvik"),"p.value"],
digits = 2)),
step_increase = 0.1)+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
labs(x = "Municipality", y = "Abundance (log EPG)", fill = "", title = "Trichostrongylidae egg abundance\nby municipality (autumn)")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))
TA16.plot.abundE <-
ggplot(TA16.sub, aes(MunicipalityDensity_Mean2016_18, ProtostrongylideLarverLPG))+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
# geom_point(fill = "gold", shape = 21, color = "black", size = 3)+
geom_point(aes(fill = Municipality2), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.height = 0.1, jitter.width = 0.05))+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
scale_x_continuous(limits = c(0.2,0.7))+
labs(x = "Moose density (mean 2016-2018)", y = "Abundance (log LPG)", fill = "")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))
TA16.plot.abundF <-
ggplot(TA16.sub, aes(Municipality2, ProtostrongylideLarverLPG))+
geom_violin(aes(fill = Municipality2), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Municipality2), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
geom_signif(comparisons = list(c("Selbu","Stjørdal")),
annotations = paste0("p = ",
round(TA16.abund.r4.posthoc.Protostrongylide.summary[which(TA16.abund.r4.posthoc.Protostrongylide.summary$covariates == "Stjørdal - Selbu"),"p.value"],
digits = 2)),
step_increase = 0.1)+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
labs(x = "Municipality", y = "Abundance (log LPG)", fill = "", title = "Protostrongylidae larvae abundance\nby municipality (autumn)")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))plot_grid(TA16.plot.abundD, TA16.plot.abundF, TA16.plot.abundC, TA16.plot.abundE,
align = "hv", axis = "tblr",
nrow = 2, labels = c("a","b","c","d"))
ggsave("figures/FigS5_Abundance_by_Municipality_autumn.png",units = "in", dpi = 1200, width = 10, height = 7)Richness
Number of parasite groups, excluding hatched strongylide (as already counted in Strongylide). All cases of ProtostrongylideLarver detection have E. alces or V. alces identified, using instead of Protostrongylide detection.
# also excluding mites because so few anyway, and we don't know if they are soil/food or host parasites
# all cases of ProtostrongylideLarver detection have E or V identified, use instead of P detection
# table(TA16[,c("EalcesLarvae01","ValcesLarvae01","ProtostrongylideLarver01")])
TA16.sub$num.of.parasite.groups <- rowSums(TA16.sub[,c("Eimeria01", "Strongylidetypeegg01", "Moniezia01", "NematodirusNematodirellaEgg01",
"Trichuris01", "Capillaria01", "Strongyloides01",
"EalcesLarvae01","ValcesLarvae01", "DictyocaulusLarver01")], na.rm = T)Using Poisson distribution, as richness is count data.
## MAIN: Exact Age and MunicipalityDensity_Mean2016_18 to reduce comparisons in model
TA16.richP.r2 <- glm(
formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
data = TA16.sub,
family = poisson(link = "log"),
na.action = na.omit
)
## SUPP:
# kg continuous
TA16.richP.r1 <- glm(
formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_kg + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
data = TA16.sub,
family = poisson(link = "log"),
na.action = na.omit
)
# age group
# TA16.richP.r2 <- glm(
# formula = num.of.parasite.groups ~ Adult_age_category + Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18,
# data = TA16.sub,
# family = poisson(link = "log"),
# na.action = na.omit
# )
# 2016 density
TA16.richP.r3 <- glm(
formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_category + Sex + MunicipalityDensity_2016 + Days.harvested.vs.lab,
data = TA16.sub,
family = poisson(link = "log"),
na.action = na.omit
)
# Municipality
TA16.richP.r4 <- glm(
formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_category + Sex + Municipality + Days.harvested.vs.lab,
data = TA16.sub,
family = poisson(link = "log"),
na.action = na.omit
)
# sex x weight interaction term
TA16.richP.r5 <- glm(
formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_category * Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
data = TA16.sub,
family = poisson(link = "log"),
na.action = na.omit
)
# age x kg cat interaction term
TA16.richP.r6 <- glm(
formula = num.of.parasite.groups ~ Exact_age * Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
data = TA16.sub,
family = poisson(link = "log"),
na.action = na.omit
)summary(TA16.richP.r1)##
## Call:
## glm(formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_kg +
## Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
## family = poisson(link = "log"), data = TA16.sub, na.action = na.omit)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.557357 0.459165 3.392 0.000695 ***
## Exact_age -0.048539 0.033538 -1.447 0.147821
## Carcass_mass_kg -0.003765 0.002708 -1.390 0.164382
## SexFemale -0.130659 0.129239 -1.011 0.312023
## MunicipalityDensity_Mean2016_18 -0.346338 0.521487 -0.664 0.506604
## Days.harvested.vs.lab 0.017513 0.014328 1.222 0.221599
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 69.719 on 116 degrees of freedom
## Residual deviance: 54.680 on 111 degrees of freedom
## AIC: 374.42
##
## Number of Fisher Scoring iterations: 4performance::check_overdispersion(TA16.richP.r1)## # Overdispersion test
##
## dispersion ratio = 0.483
## Pearson's Chi-Squared = 53.563
## p-value = 1## No overdispersion detected.summary(TA16.richP.r2)##
## Call:
## glm(formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_category +
## Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
## family = poisson(link = "log"), data = TA16.sub, na.action = na.omit)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.07136 0.28189 3.801 0.000144 ***
## Exact_age -0.07498 0.02933 -2.556 0.010585 *
## Carcass_mass_category.L 0.04098 0.10801 0.379 0.704364
## SexFemale -0.10811 0.12921 -0.837 0.402765
## MunicipalityDensity_Mean2016_18 -0.25524 0.52439 -0.487 0.626444
## Days.harvested.vs.lab 0.01748 0.01432 1.221 0.221960
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 69.719 on 116 degrees of freedom
## Residual deviance: 56.486 on 111 degrees of freedom
## AIC: 376.23
##
## Number of Fisher Scoring iterations: 5performance::check_overdispersion(TA16.richP.r2)## # Overdispersion test
##
## dispersion ratio = 0.497
## Pearson's Chi-Squared = 55.159
## p-value = 1## No overdispersion detected.summary(TA16.richP.r3)##
## Call:
## glm(formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_category +
## Sex + MunicipalityDensity_2016 + Days.harvested.vs.lab, family = poisson(link = "log"),
## data = TA16.sub, na.action = na.omit)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.06572 0.29840 3.571 0.000355 ***
## Exact_age -0.07495 0.02936 -2.552 0.010696 *
## Carcass_mass_category.L 0.03826 0.10732 0.357 0.721457
## SexFemale -0.10706 0.12924 -0.828 0.407442
## MunicipalityDensity_2016 -0.24510 0.56833 -0.431 0.666273
## Days.harvested.vs.lab 0.01766 0.01431 1.234 0.217195
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 69.719 on 116 degrees of freedom
## Residual deviance: 56.537 on 111 degrees of freedom
## AIC: 376.28
##
## Number of Fisher Scoring iterations: 5performance::check_overdispersion(TA16.richP.r3)## # Overdispersion test
##
## dispersion ratio = 0.497
## Pearson's Chi-Squared = 55.194
## p-value = 1## No overdispersion detected.summary(TA16.richP.r4)##
## Call:
## glm(formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_category +
## Sex + Municipality + Days.harvested.vs.lab, family = poisson(link = "log"),
## data = TA16.sub, na.action = na.omit)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.81666 0.26541 3.077 0.00209 **
## Exact_age -0.07657 0.02942 -2.603 0.00925 **
## Carcass_mass_category.L 0.04874 0.10933 0.446 0.65575
## SexFemale -0.12429 0.13061 -0.952 0.34130
## MunicipalityMeråker -0.08717 0.37502 -0.232 0.81620
## MunicipalitySelbu 0.18860 0.26603 0.709 0.47836
## MunicipalityStjørdal 0.20193 0.34400 0.587 0.55720
## MunicipalityTydal 0.38892 0.32263 1.205 0.22802
## Days.harvested.vs.lab 0.01336 0.01474 0.906 0.36468
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 69.719 on 116 degrees of freedom
## Residual deviance: 53.901 on 108 degrees of freedom
## AIC: 379.64
##
## Number of Fisher Scoring iterations: 4performance::check_overdispersion(TA16.richP.r4)## # Overdispersion test
##
## dispersion ratio = 0.484
## Pearson's Chi-Squared = 52.277
## p-value = 1## No overdispersion detected.summary(TA16.richP.r5)##
## Call:
## glm(formula = num.of.parasite.groups ~ Exact_age + Carcass_mass_category *
## Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
## family = poisson(link = "log"), data = TA16.sub, na.action = na.omit)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.040208 0.292469 3.557 0.000376 ***
## Exact_age -0.074721 0.029316 -2.549 0.010809 *
## Carcass_mass_category.L -0.008176 0.163693 -0.050 0.960166
## SexFemale -0.074577 0.153991 -0.484 0.628176
## MunicipalityDensity_Mean2016_18 -0.242498 0.525862 -0.461 0.644696
## Days.harvested.vs.lab 0.017682 0.014322 1.235 0.216978
## Carcass_mass_category.L:SexFemale 0.086307 0.212267 0.407 0.684306
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 69.719 on 116 degrees of freedom
## Residual deviance: 56.320 on 110 degrees of freedom
## AIC: 378.06
##
## Number of Fisher Scoring iterations: 5performance::check_overdispersion(TA16.richP.r5)## # Overdispersion test
##
## dispersion ratio = 0.501
## Pearson's Chi-Squared = 55.086
## p-value = 1## No overdispersion detected.summary(TA16.richP.r6)##
## Call:
## glm(formula = num.of.parasite.groups ~ Exact_age * Carcass_mass_category +
## Sex + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
## family = poisson(link = "log"), data = TA16.sub, na.action = na.omit)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.077567 0.283391 3.802 0.000143 ***
## Exact_age -0.078471 0.034244 -2.292 0.021934 *
## Carcass_mass_category.L 0.058978 0.137951 0.428 0.668991
## SexFemale -0.109125 0.129261 -0.844 0.398545
## MunicipalityDensity_Mean2016_18 -0.250292 0.524664 -0.477 0.633325
## Days.harvested.vs.lab 0.017249 0.014364 1.201 0.229808
## Exact_age:Carcass_mass_category.L -0.009813 0.047298 -0.207 0.835637
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 69.719 on 116 degrees of freedom
## Residual deviance: 56.442 on 110 degrees of freedom
## AIC: 378.19
##
## Number of Fisher Scoring iterations: 5performance::check_overdispersion(TA16.richP.r6)## # Overdispersion test
##
## dispersion ratio = 0.499
## Pearson's Chi-Squared = 54.911
## p-value = 1## No overdispersion detected.plot(allEffects(TA16.richP.r1))
plot(allEffects(TA16.richP.r2))
Model selection
TA16.richP.aic <- AIC(TA16.richP.r1, TA16.richP.r2, TA16.richP.r3, TA16.richP.r4, TA16.richP.r5, TA16.richP.r6)
TA16.richP.aic[order(TA16.richP.aic$AIC),]## df AIC
## TA16.richP.r1 6 374.4226
## TA16.richP.r2 6 376.2291
## TA16.richP.r3 6 376.2801
## TA16.richP.r5 7 378.0627
## TA16.richP.r6 7 378.1850
## TA16.richP.r4 9 379.6437# r1, then r2c, then r3 Fig. S2ab
TA16.plot.richA <-
ggplot(subset(TA16.sub, !is.na(num.of.parasite.groups)),
aes(Exact_age, num.of.parasite.groups))+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
geom_point(shape = 21, color = "black", fill = "green3", size = 3)+
# geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3)+
# scale_fill_manual(values = c("#C7E9C0","#74C476","#006D2C"))+
scale_y_continuous(limits = c(0, round(max(TA16.sub$num.of.parasite.groups, na.rm = T)+1)))+
labs(x = "\nAge (years)", y = "No. of parasite groups", fill = "", title = "Autumn\n")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))
TA16.plot.richB <-
ggplot(subset(TA16.sub, !is.na(num.of.parasite.groups)),
aes(Carcass_mass_category, num.of.parasite.groups))+
# geom_boxplot(aes(fill = Carcass_mass_category), outlier.colour = NA, alpha = 0.25)+
geom_violin(aes(fill = Carcass_mass_category), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Carcass_mass_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#3C2692","#CBC9DA"))+
scale_y_continuous(limits = c(0, round(max(TA16.sub$num.of.parasite.groups, na.rm = T)+1)))+
labs(x = "\nCarcass mass category", y = "No. of parasite groups", fill = "", title = "Autumn\n")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))TA16.plot.richA## `geom_smooth()` using formula = 'y ~ x'
TA16.plot.richB
Other figures
ggplot(subset(TA16.sub, !is.na(num.of.parasite.groups)),
aes(Adult_age_category, num.of.parasite.groups))+
geom_boxplot(aes(fill = Sex), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Sex), shape = 21, color = "black", size = 2,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#E41A1C","#377EB8"))+
scale_x_discrete(labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
labs(x = "Age group", y = "No. of parasite groups", fill = "Sex")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
ggplot(subset(TA16.sub, !is.na(num.of.parasite.groups)),
aes(Municipality, num.of.parasite.groups))+
geom_boxplot(aes(fill = Municipality), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 2,
position = position_jitterdodge(jitter.width = 0.2))+
labs(x = "Municipality", y = "No. of parasite groups", fill = "")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
ggplot(subset(TA16.sub, !is.na(num.of.parasite.groups)),
aes(MunicipalityDensity_2016, num.of.parasite.groups))+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 2)+
labs(x = "Harvest density (2016)", y = "No. of parasite groups", fill = "Municipality")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
ggplot(subset(TA16.sub, !is.na(num.of.parasite.groups)),
aes(MunicipalityDensity_Mean2016_18, num.of.parasite.groups))+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 2)+
labs(x = "Harvest density (2016-2018)", y = "No. of parasite groups", fill = "Municipality")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
Table 3
Stat tables
# main table, present best AIC fit
TA16.richP.r1.summary <- summary(TA16.richP.r1)
colnames(TA16.richP.r1.summary$coefficients) <- c("estimate","std.error","statistic","p.value")
TA16.richP.r1.summary <- data.frame(
resp_var="num.of.parasite.groups",
covariates=row.names(TA16.richP.r1.summary$coefficients),
as.data.frame(TA16.richP.r1.summary$coefficients),
p.adj=NA,
row.names = NULL
)
TA16.prev.r2$summary[which(TA16.prev.r2$summary$covariates != "(Intercept)" & TA16.prev.r2$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 26 ProtostrongylideLarver01 Exact_age -2.2002240 0.5571874
## 32 EalcesLarvae01 Exact_age -3.3206061 0.8611806
## 33 EalcesLarvae01 Carcass_mass_category.L 1.7491054 0.8007562
## 38 ValcesLarvae01 Exact_age -1.3140766 0.4485914
## 39 ValcesLarvae01 Carcass_mass_category.L 0.9670895 0.4926710
## statistic p.value p.adj n.pos
## 26 -3.948805 7.854244e-05 0.00242163 61
## 32 -3.855876 1.153157e-04 0.00242163 53
## 33 2.184317 2.893896e-02 0.15192952 53
## 38 -2.929340 3.396827e-03 0.02419897 48
## 39 1.962952 4.965178e-02 0.20853749 48TA16.abund.r4$summary[which(TA16.abund.r4$summary$resp_var %in% c("StrongylidetypeeggEPG","ProtostrongylideLarverLPG") & TA16.abund.r4$summary$covariates != "(Intercept)" & TA16.abund.r4$summary$p.value < 0.05),]## resp_var covariates estimate std.error
## 3 StrongylidetypeeggEPG Carcass_mass_category.L 0.23694734 0.12070813
## 4 StrongylidetypeeggEPG SexFemale -0.35246577 0.16050113
## 6 StrongylidetypeeggEPG MunicipalitySelbu 0.90794612 0.33869841
## 29 ProtostrongylideLarverLPG Exact_age -1.91127716 0.39672676
## 30 ProtostrongylideLarverLPG Carcass_mass_category.L 0.79327464 0.17718616
## 31 ProtostrongylideLarverLPG SexFemale -0.57447542 0.24157287
## 34 ProtostrongylideLarverLPG MunicipalityStjørdal 1.55677808 0.69726554
## 36 ProtostrongylideLarverLPG Days.harvested.vs.lab 0.09055432 0.02434742
## statistic p.value p.adj n.pos
## 3 1.962977 4.964879e-02 1.378070e-01 114
## 4 -2.196033 2.808958e-02 9.192955e-02 114
## 6 2.680692 7.347011e-03 3.778463e-02 114
## 29 -4.817616 1.452837e-06 1.743404e-05 61
## 30 4.477069 7.567481e-06 6.810733e-05 61
## 31 -2.378063 1.740387e-02 6.961547e-02 61
## 34 2.232690 2.556937e-02 9.192955e-02 61
## 36 3.719257 1.998095e-04 1.438628e-03 61TA16.richP.r1.summary[which(TA16.richP.r1.summary$covariates != "(Intercept)" & TA16.richP.r1.summary$p.value < 0.05),]## [1] resp_var covariates estimate std.error statistic p.value p.adj
## <0 rows> (or 0-length row.names)stats.tab1 <- rbind.data.frame(
data.frame(model="M2", TA16.prev.r2$summary[which(TA16.prev.r2$summary$resp_var %in% c("ProtostrongylideLarver01","EalcesLarvae01","ValcesLarvae01")),]),
data.frame(model="M4", TA16.abund.r4$summary[which(TA16.abund.r4$summary$resp_var %in% c("StrongylidetypeeggEPG") &
TA16.abund.r4$summary$covariates %in% c("(Intercept)","Exact_age","Carcass_mass_category.L","SexFemale","Days.harvested.vs.lab")),]),
data.frame(model="M4", TA16.abund.r4.posthoc.Strongylide.summary[which(TA16.abund.r4.posthoc.Strongylide.summary$p.value < 0.1),], n.pos=""), # put rest in supp
data.frame(model="M4", TA16.abund.r4$summary[which(TA16.abund.r4$summary$resp_var %in% c("ProtostrongylideLarverLPG") &
TA16.abund.r4$summary$covariates %in% c("(Intercept)","Exact_age","Carcass_mass_category.L","SexFemale","Days.harvested.vs.lab")),]),
data.frame(model="M4", TA16.abund.r4.posthoc.Protostrongylide.summary[which(TA16.abund.r4.posthoc.Protostrongylide.summary$p.value < 0.1),], n.pos=""), # put rest in supp
data.frame(model="M1", TA16.richP.r1.summary, n.pos=length(which(!is.na(TA16.sub$num.of.parasite.groups))))
)
stats.tab1$coeff_print <- ifelse(is.na(stats.tab1$estimate),"",paste(round(stats.tab1$estimate, digits = 3),"\u00B1", round(stats.tab1$std.error, digits = 3)))
stats.tab1$pval_print <- ifelse(is.na(stats.tab1$p.value),"",ifelse(stats.tab1$p.value >= 0.001, round(stats.tab1$p.value, digits = 3), "<0.001"))
stats.tab1$padj_print <- ifelse(is.na(stats.tab1$p.adj),"",ifelse(stats.tab1$p.adj >= 0.001, round(stats.tab1$p.adj, digits = 3), "<0.001"))
stats.tab1[,c("model","resp_var","n.pos","covariates","coeff_print","pval_print","padj_print")]## model resp_var n.pos covariates
## 25 M2 ProtostrongylideLarver01 61 (Intercept)
## 26 M2 ProtostrongylideLarver01 61 Exact_age
## 27 M2 ProtostrongylideLarver01 61 Carcass_mass_category.L
## 28 M2 ProtostrongylideLarver01 61 SexFemale
## 29 M2 ProtostrongylideLarver01 61 MunicipalityDensity_Mean2016_18
## 30 M2 ProtostrongylideLarver01 61 Days.harvested.vs.lab
## 31 M2 EalcesLarvae01 53 (Intercept)
## 32 M2 EalcesLarvae01 53 Exact_age
## 33 M2 EalcesLarvae01 53 Carcass_mass_category.L
## 34 M2 EalcesLarvae01 53 SexFemale
## 35 M2 EalcesLarvae01 53 MunicipalityDensity_Mean2016_18
## 36 M2 EalcesLarvae01 53 Days.harvested.vs.lab
## 37 M2 ValcesLarvae01 48 (Intercept)
## 38 M2 ValcesLarvae01 48 Exact_age
## 39 M2 ValcesLarvae01 48 Carcass_mass_category.L
## 40 M2 ValcesLarvae01 48 SexFemale
## 41 M2 ValcesLarvae01 48 MunicipalityDensity_Mean2016_18
## 42 M2 ValcesLarvae01 48 Days.harvested.vs.lab
## 1 M4 StrongylidetypeeggEPG 114 (Intercept)
## 2 M4 StrongylidetypeeggEPG 114 Exact_age
## 3 M4 StrongylidetypeeggEPG 114 Carcass_mass_category.L
## 4 M4 StrongylidetypeeggEPG 114 SexFemale
## 9 M4 StrongylidetypeeggEPG 114 Days.harvested.vs.lab
## 21 M4 StrongylidetypeeggEPG Selbu - Malvik
## 281 M4 ProtostrongylideLarverLPG 61 (Intercept)
## 291 M4 ProtostrongylideLarverLPG 61 Exact_age
## 301 M4 ProtostrongylideLarverLPG 61 Carcass_mass_category.L
## 311 M4 ProtostrongylideLarverLPG 61 SexFemale
## 361 M4 ProtostrongylideLarverLPG 61 Days.harvested.vs.lab
## 8 M4 ProtostrongylideLarverLPG Stjørdal - Selbu
## 11 M1 num.of.parasite.groups 117 (Intercept)
## 22 M1 num.of.parasite.groups 117 Exact_age
## 310 M1 num.of.parasite.groups 117 Carcass_mass_kg
## 43 M1 num.of.parasite.groups 117 SexFemale
## 5 M1 num.of.parasite.groups 117 MunicipalityDensity_Mean2016_18
## 6 M1 num.of.parasite.groups 117 Days.harvested.vs.lab
## coeff_print pval_print padj_print
## 25 5.588 ± 1.612 <0.001 0.007
## 26 -2.2 ± 0.557 <0.001 0.002
## 27 1.13 ± 0.675 0.094 0.304
## 28 -0.762 ± 0.586 0.193 0.406
## 29 -3.805 ± 2.633 0.149 0.367
## 30 0.124 ± 0.083 0.138 0.367
## 31 5.227 ± 1.666 0.002 0.018
## 32 -3.321 ± 0.861 <0.001 0.002
## 33 1.749 ± 0.801 0.029 0.152
## 34 -1.005 ± 0.592 0.09 0.304
## 35 -0.226 ± 2.589 0.931 0.996
## 36 0.118 ± 0.084 0.163 0.38
## 37 3.79 ± 1.296 0.003 0.024
## 38 -1.314 ± 0.449 0.003 0.024
## 39 0.967 ± 0.493 0.05 0.209
## 40 -0.788 ± 0.504 0.118 0.354
## 41 -3.835 ± 2.244 0.088 0.304
## 42 0.081 ± 0.062 0.188 0.406
## 1 4.738 ± 0.348 <0.001 <0.001
## 2 -0.003 ± 0.024 0.901 0.927
## 3 0.237 ± 0.121 0.05 0.138
## 4 -0.352 ± 0.161 0.028 0.092
## 9 -0.039 ± 0.02 0.052 0.138
## 21 0.908 ± 0.339 0.052
## 281 4.134 ± 0.753 <0.001 <0.001
## 291 -1.911 ± 0.397 <0.001 <0.001
## 301 0.793 ± 0.177 <0.001 <0.001
## 311 -0.574 ± 0.242 0.017 0.07
## 361 0.091 ± 0.024 <0.001 0.001
## 8 1.051 ± 0.359 0.025
## 11 1.557 ± 0.459 <0.001
## 22 -0.049 ± 0.034 0.148
## 310 -0.004 ± 0.003 0.164
## 43 -0.131 ± 0.129 0.312
## 5 -0.346 ± 0.521 0.507
## 6 0.018 ± 0.014 0.222write.table(stats.tab1[,c("resp_var","model","n.pos","covariates","coeff_print","pval_print","padj_print")],
"figures/table_statistic_models_autumn.txt",
row.names = F, quote = F, sep = "\t", fileEncoding = "latin1")Autumn vs Winter Trøndelag
Associations comparing adult moose (female, due to low male numbers) between autumn 2016 (hunted) and winter 2017 (GPS-marked) in Trøndelag (code TS1617 - “S” for seasonal differences).
Summaries
Generate some basic summaries of metadata.
# subset to adult female moose in Trøndelag in 2016 and 2017
TS1617 <- dat[which(dat$Year %in% c("2017","2016") &
dat$Adult_age_category %in% c("agegroup2","agegroup3") &
dat$Sex == "Female"),]
table(TS1617[,c("Adult_age_category","Year")])## Year
## Adult_age_category 2016 2017
## agegroup1 0 0
## agegroup2 25 14
## agegroup3 17 6TS1617 <- TS1617[which(!is.na(TS1617$Municipality)),]Table 1b
Basic stats
table(TS1617$McMaster_performed)##
## 1
## 62table(TS1617$Baermann_performed)##
## 0 1
## 8 54table(TS1617$Sex, TS1617$Adult_age_category, TS1617$Season)## , , = Autumn
##
##
## agegroup1 agegroup2 agegroup3
## Male 0 0 0
## Female 0 25 17
##
## , , = Winter
##
##
## agegroup1 agegroup2 agegroup3
## Male 0 0 0
## Female 0 14 6table(TS1617$Sex, TS1617$Season)##
## Autumn Winter
## Male 0 0
## Female 42 20sum(table(TS1617$Sex, TS1617$Adult_age_category))## [1] 62# summary table
ms.tab2 <- as.data.frame(table(TS1617$Sex, TS1617$Adult_age_category, TS1617$Season))
names(ms.tab2) <- c("Sex","Adult_age_category","Season","count")
ms.tab2 <- rbind.data.frame(
ms.tab2,
data.frame(Sex="Female", Adult_age_category="all", Season="all", count=nrow(TS1617))
)
ms.tab2$Carcass_mass_kg.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "Carcass_mass_kg",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617)
# add major parasite groups
# same as abundance/prevalence models, must be at least 5 positive samples to include in table
# v1 - intensity
ms.tab2$StrongylidetypeeggEPG.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "StrongylidetypeeggEPG",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617, intensity = TRUE)
ms.tab2$NematodirussppEPG.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "NematodirussppEPG",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617, intensity = TRUE)
ms.tab2$NematodirusBattusEPG.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "NematodirusBattusEPG",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617, intensity = TRUE)
ms.tab2$TrichurisEPG.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "TrichurisEPG",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617, intensity = TRUE)
ms.tab2$CapillariaEPG.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "CapillariaEPG",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617, intensity = TRUE)
ms.tab2$EimeriaOPG.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "EimeriaOPG",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617, intensity = TRUE)
ms.tab2$ProtostrongylideLarverLPG.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "ProtostrongylideLarverLPG",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617, intensity = TRUE)
ms.tab2$DictyocaulusLPG.Median <- sapply(seq(1,nrow(ms.tab2)), calc_median_IQR_label, v = "DictyocaulusLPG",
groups = c("Sex","Adult_age_category","Season"),
sum.tab = ms.tab2, df.meta = TS1617, intensity = TRUE)
ms.tab2$Carcass_mass_category <- NA Table 2b
Autumn vs winter, summary table of prevalence by age and season
TS1617.det.df <- melt(
TS1617[,c("IndividualID","Season","Adult_age_category",
"Municipality","MunicipalityDensity_2016","MunicipalityDensity_Mean2016_18","McMaster_performed","Baermann_performed",
mcmaster.parasite.groups.01, baermann.parasite.groups.01)],
id.vars = c("IndividualID","Season","Adult_age_category",
"Municipality","MunicipalityDensity_2016","MunicipalityDensity_Mean2016_18","McMaster_performed","Baermann_performed"),
measure.vars = c(mcmaster.parasite.groups.01, baermann.parasite.groups.01)
)
TS1617.det.by.group <- dcast(
TS1617.det.df,
Season + Adult_age_category ~ variable,
value.var = "value",
fun.aggregate = sum, fill = 0, drop = T, na.rm = T
)
## Calculate no. of individuals based on samples for which test was performed
TS1617.det.by.group$num_inds.MM <- sapply(seq(1,nrow(TS1617.det.by.group)), function(i){
s <- TS1617.det.by.group[i,"Season"]
age <- TS1617.det.by.group[i,"Adult_age_category"]
length(which(TS1617$Season == s & TS1617$Adult_age_category == age & TS1617$McMaster_performed == 1))
})
TS1617.det.by.group$num_inds.BA <- sapply(seq(1,nrow(TS1617.det.by.group)), function(i){
s <- TS1617.det.by.group[i,"Season"]
age <- TS1617.det.by.group[i,"Adult_age_category"]
length(which(TS1617$Season == s & TS1617$Adult_age_category == age & TS1617$Baermann_performed == 1))
})
TS1617.det.by.group.hm <- melt(TS1617.det.by.group,
id.vars = c("Season","Adult_age_category","num_inds.MM","num_inds.BA"),
measure.vars = c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
variable.name = "parasite", value.name = "n.positive")
TS1617.det.by.group.hm$group <- paste(TS1617.det.by.group.hm$Season, TS1617.det.by.group.hm$Adult_age_category, sep = "_")
TS1617.det.by.group.hm$proportion <- ifelse(
TS1617.det.by.group.hm$parasite %in% mcmaster.parasite.groups.01,
TS1617.det.by.group.hm$n.positive / TS1617.det.by.group.hm$num_inds.MM,
TS1617.det.by.group.hm$n.positive / TS1617.det.by.group.hm$num_inds.BA
)
# add 95%CI to prevalence
TS1617.det.by.group.hm$CI95.lower <- sapply(seq(1,nrow(TS1617.det.by.group.hm)), function(i){
p <- as.character(TS1617.det.by.group.hm[i,"parasite"])
n.det <- TS1617.det.by.group.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TS1617.det.by.group.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TS1617.det.by.group.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[1])
})
TS1617.det.by.group.hm$CI95.upper <- sapply(seq(1,nrow(TS1617.det.by.group.hm)), function(i){
p <- as.character(TS1617.det.by.group.hm[i,"parasite"])
n.det <- TS1617.det.by.group.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TS1617.det.by.group.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TS1617.det.by.group.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[2])
})
TS1617.det.by.group.hm$parasite <- factor(TS1617.det.by.group.hm$parasite,
levels = c("MiddEgg01", "Moniezia01", "Eimeria01",
"Strongyloides01", "Capillaria01", "Trichuris01", "NematodirusNematodirellaEgg01",
"Strongylidetypeegg01",
"ValcesLarvae01", "EalcesLarvae01", "ProtostrongylideLarver01",
"DictyocaulusLarver01"))
# exclude mites
TS1617.det.by.group.hm <- TS1617.det.by.group.hm[which(TS1617.det.by.group.hm$parasite != "MiddEgg01"),]
TS1617.det.by.group.hm$parasite <- droplevels(TS1617.det.by.group.hm$parasite)TS1617.det.by.group.hm$percent <- round(TS1617.det.by.group.hm$proportion * 100, 1)
TS1617.det.by.group.hm$label_v2 <- sapply(seq(1,nrow(TS1617.det.by.group.hm)), function(i){
ct <- TS1617.det.by.group.hm[i, "n.positive"]
if(ct == 0){
return(0)
} else {
per <- TS1617.det.by.group.hm[i,"percent"]
return(paste0(ct," (",per,"%)"))
}
})
TS1617.det.by.group.hm$label_v3 <- sapply(seq(1,nrow(TS1617.det.by.group.hm)), function(i){
if(TS1617.det.by.group.hm[i,"proportion"] == 0){
return(0)
} else {
prop <- round(TS1617.det.by.group.hm[i,"proportion"], digits = 2)
ciL <- round(TS1617.det.by.group.hm[i,"CI95.lower"], digits = 1)
ciU <- round(TS1617.det.by.group.hm[i,"CI95.upper"], digits = 1)
return(paste0(prop,"\n(",ciL,"-",ciU,")"))
}
})
TS1617.det.by.group[,c("Season","Adult_age_category","num_inds.BA","num_inds.MM")]## Season Adult_age_category num_inds.BA num_inds.MM
## 1 Autumn agegroup2 20 25
## 2 Autumn agegroup3 14 17
## 3 Winter agegroup2 14 14
## 4 Winter agegroup3 6 6prev.fig2 <-
ggplot(TS1617.det.by.group.hm, aes(group, parasite, fill = proportion))+
geom_tile(colour = "grey50", linewidth = 0.5)+
scale_fill_gradient(low="white", high="grey50")+
# geom_vline(xintercept = 2.5, color = "black", linewidth = 0.5, linetype = 2)+
geom_text(aes(label=label_v3))+
scale_y_discrete(labels = c("*Moniezia* spp. eggs","*Eimeria* spp. oocysts",
"*Strongyloides* spp. eggs","*Capillaria* spp. eggs","*Trichuris* spp. eggs","Nematodirinae eggs",
"Trichostrongylidae spp.", "*V. alces* larvae","*E. alces* larvae","Protostrongylidae larvae",
"*Dictyocaulus* spp. larvae"))+
scale_x_discrete(labels = c("Autumn\nFemale\n2-5 yr\nnL=20\nnE=25","Autumn\nFemale\n6+ yr\nnL=14\nnE=17",
"Winter\nFemale\n2-5 yr\nnL=14\nnE=14","Winter\nFemale\n6+ yr\nnL=6\nnE=6"),
position = "top")+
labs(y = "", x = "", fill = "Prevalence \n")+
theme_bw()+
guides(fill = guide_colorbar(barwidth = 10, barheight = 1))+
theme(axis.title = element_text(size = 13, colour = "black"),
# axis.text.y = ggtext::element_markdown(size = 13, colour = "black", hjust = 1),
axis.text.y = element_blank(),
axis.text.x = element_text(size = 12, colour = "black", angle = 0, vjust = 0.5),
legend.text = element_text(size = 11, colour = "black"), legend.title = element_text(size = 12, colour = "black"),
legend.position = "none",
panel.border = element_blank(), axis.ticks = element_blank())prev.fig2
prop.table(table(TS1617$Strongylidetypeegg01, TS1617$Season), margin = 2)##
## Autumn Winter
## 0 0.07142857 0.10000000
## 1 0.92857143 0.90000000Other figures
TS1617.abund.df <- melt(
TS1617[,c("IndividualID","Season","Adult_age_category",
"Municipality","MunicipalityDensity_2016","MunicipalityDensity_Mean2016_18","McMaster_performed","Baermann_performed",
mcmaster.parasite.groups.EPG, baermann.parasite.groups.LPG)],
id.vars = c("IndividualID","Season","Adult_age_category",
"Municipality","MunicipalityDensity_2016","MunicipalityDensity_Mean2016_18","McMaster_performed","Baermann_performed"),
measure.vars = c(mcmaster.parasite.groups.EPG, baermann.parasite.groups.LPG)
)ggplot(subset(TS1617.abund.df, !variable %in% c("EimeriaOPG","CapillariaEPG","MiddEggEPG",
"NematodirusBattusEPG","DictyocaulusLPG")),
aes(Season, value))+
geom_violin(aes(fill = Adult_age_category), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#74C476","#006D2C"), labels = c(agegroup1="1 yr",agegroup2="2-5 yr",agegroup3="6+ yr"))+
# scale_x_discrete()+
# scale_y_continuous(transform = "log10")+
labs(x = "Season", y = "Abundance (log scale)", fill = "Age group", title = "")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable, nrow = 1, scales = "free_y")
Models
GLMs testing for associations between season (factor) and age group (ordered factor), as well as environment (mean harvest density over 2016-2018, to match Autumn models). Includes adjustment for multiple hypothesis testing.
Must be present in at least 5 samples to test (otherwise not enough variation in dataset). One model per parasite:
- season + age group + harvest density (mean over 2016-2018)
Prevalence
## MAIN
TS1617.prev <- run_glm(
TS1617,
c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
" ~ Season + Adult_age_category + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "binomial",
min.samples = 5
)## [1] "Skipping Eimeria01, not enough positive samples."
## [1] "Running logistic model for Strongylidetypeegg01"
## [1] "Running logistic model for NematodirusNematodirellaEgg01"
## [1] "Running logistic model for Trichuris01"
## [1] "Skipping Capillaria01, not enough positive samples."
## [1] "Skipping MiddEgg01, not enough positive samples."
## [1] "Skipping Moniezia01, not enough positive samples."
## [1] "Skipping Strongyloides01, not enough positive samples."
## [1] "Running logistic model for ProtostrongylideLarver01"
## [1] "Skipping EalcesLarvae01, not enough positive samples."
## [1] "Skipping ValcesLarvae01, not enough positive samples."
## [1] "Skipping DictyocaulusLarver01, not enough positive samples."TS1617.prev$summary[which(TS1617.prev$summary$covariates != "(Intercept)" & TS1617.prev$summary$p.value < 0.1),]## resp_var covariates estimate std.error statistic
## 3 Strongylidetypeegg01 Adult_age_category.L -1.519379 0.8426482 -1.8031
## p.value p.adj
## 3 0.07137245 0.5976107Add no. of positive samples to table
TS1617.prev$summary$n.pos <- sapply(TS1617.prev$summary$resp_var, function(x){
sum(TS1617[,x], na.rm = T)
})ggplot(subset(TS1617.det.df, variable %in% c("Strongylidetypeegg01")),
aes(Season, value))+
geom_violin(aes(fill = Adult_age_category), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2, jitter.height = 0.05))+
scale_fill_manual(values = c("#74C476","#006D2C"), labels = c(agegroup2="2-5 yr",agegroup3="6+ yr"))+
scale_x_discrete(labels = c("Autumn"="Autumn (2016)","Winter"="Winter (2017)"))+
labs(x = "Season (year)", y = "Detection", fill = "Age group")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(Strongylidetypeegg01="Trichostrongylidae eggs")))
Abundance
Rounding abundances to nearest integer and running negative binomial GLMs.
## MAIN
TS1617.abund <- run_glm(
TS1617,
c(mcmaster.parasite.groups.EPG,baermann.parasite.groups.LPG),
" ~ Season + Adult_age_category + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab",
fam = "nbinom1",
ct_round = T,
min.samples = 5
)## [1] "Skipping EimeriaOPG, not enough positive samples."
## [1] "Running Neg Binomial model for StrongylidetypeeggEPG"
## [1] "Running Neg Binomial model for NematodirussppEPG"
## [1] "Skipping NematodirusBattusEPG, not enough positive samples."
## [1] "Running Neg Binomial model for TrichurisEPG"
## [1] "Skipping CapillariaEPG, not enough positive samples."
## [1] "Skipping MiddEggEPG, not enough positive samples."
## [1] "Running Neg Binomial model for ProtostrongylideLarverLPG"
## [1] "Skipping DictyocaulusLPG, not enough positive samples."TS1617.abund$summary[which(TS1617.abund$summary$covariates != "(Intercept)" & TS1617.abund$summary$p.value < 0.1),]## resp_var covariates estimate std.error
## 3 StrongylidetypeeggEPG Adult_age_category.L -0.3246267 0.1858473
## 14 TrichurisEPG MunicipalityDensity_Mean2016_18 -3.5752370 2.1179319
## statistic p.value p.adj
## 3 -1.746739 0.08068262 0.3560971
## 14 -1.688079 0.09139602 0.3560971Add no. of positive samples to table
TS1617.abund$summary$n.pos <- sapply(TS1617.abund$summary$resp_var, function(x){
sum(TS1617[,x] > 0, na.rm = T)
})Fig. 4
TS1617.plot.detA <-
ggplot(subset(TS1617.abund.df, variable %in% c("StrongylidetypeeggEPG")),
aes(Season, value))+
geom_boxplot(aes(fill = Season), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Season), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c(Autumn="#E57200",Winter="#004F71"))+
scale_x_discrete(labels = c(Autumn="Autumn\n2016",Winter="Winter\n2017"))+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
labs(x = "Season", y = "Abundance (log EPG)", fill = "Season")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(StrongylidetypeeggEPG="Trichostrongylidae")))TS1617.det.by.season <- dcast(
TS1617.det.df,
Season ~ variable,
value.var = "value",
fun.aggregate = sum, fill = 0, drop = T, na.rm = T
)
## Calculate no. of individuals based on samples for which test was performed
TS1617.det.by.season$num_inds.MM <- sapply(seq(1,nrow(TS1617.det.by.season)), function(i){
sn <- TS1617.det.by.season[i,"Season"]
length(which(TS1617$Season == sn & TS1617$McMaster_performed == 1))
})
TS1617.det.by.season$num_inds.BA <- sapply(seq(1,nrow(TS1617.det.by.season)), function(i){
sn <- TS1617.det.by.season[i,"Season"]
length(which(TS1617$Season == sn & TS1617$Baermann_performed == 1))
})
TS1617.det.by.season.hm <- melt(TS1617.det.by.season,
id.vars = c("Season","num_inds.MM","num_inds.BA"),
measure.vars = c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
variable.name = "parasite", value.name = "n.positive")
TS1617.det.by.season.hm$proportion <- ifelse(
TS1617.det.by.season.hm$parasite %in% mcmaster.parasite.groups.01,
TS1617.det.by.season.hm$n.positive / TS1617.det.by.season.hm$num_inds.MM,
TS1617.det.by.season.hm$n.positive / TS1617.det.by.season.hm$num_inds.BA
)
# add 95%CI to prevalence
TS1617.det.by.season.hm$CI95.lower <- sapply(seq(1,nrow(TS1617.det.by.season.hm)), function(i){
p <- as.character(TS1617.det.by.season.hm[i,"parasite"])
n.det <- TS1617.det.by.season.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TS1617.det.by.season.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TS1617.det.by.season.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[1])
})
TS1617.det.by.season.hm$CI95.upper <- sapply(seq(1,nrow(TS1617.det.by.season.hm)), function(i){
p <- as.character(TS1617.det.by.season.hm[i,"parasite"])
n.det <- TS1617.det.by.season.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TS1617.det.by.season.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TS1617.det.by.season.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[2])
})
TS1617.det.by.season.hm$CI95.lower.v2 <- ifelse(TS1617.det.by.season.hm$proportion == 0,NA,TS1617.det.by.season.hm$CI95.lower)
TS1617.det.by.season.hm$CI95.upper.v2 <- ifelse(TS1617.det.by.season.hm$proportion == 0,NA,TS1617.det.by.season.hm$CI95.upper)
TS1617.det.by.season.hm$variable <- factor(TS1617.det.by.season.hm$parasite,
levels = levels(TS1617.det.df$variable))TS1617.plot.detB <-
ggplot(
subset(TS1617.det.by.season.hm, variable %in% c("Trichuris01")))+
geom_errorbar(aes(x = Season, ymin = CI95.lower.v2, ymax = CI95.upper.v2),
width = 0.5, linewidth = 0.75, colour = "black")+
geom_point(aes(Season, proportion, fill = Season, size = num_inds.BA), shape = 21)+
scale_fill_manual(values = c(Autumn="#E57200",Winter="#004F71"), )+
scale_x_discrete(labels = c(Autumn="Autumn\n2016",Winter="Winter\n2017"))+
labs(x = "Season", y = "Prevalence", size = "Total no.\nof samples")+
guides(fill = "none")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = ggtext::element_markdown(size = 13),
legend.position = "right",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(Trichuris01="*Trichuris*")))plot_grid(TS1617.plot.detA, TS1617.plot.detB,
nrow = 1, align = "h", axis = "tb",
rel_widths = c(0.75,1),
labels = c("a","b"))
ggsave("figures/Fig4_AbundancePrevalence_by_Season.png",units = "in", dpi = 1200, width = 7.5, height = 4)ggplot(subset(TS1617.abund.df, variable %in% c("StrongylidetypeeggEPG")),
aes(Season, value))+
geom_boxplot(aes(fill = Adult_age_category), alpha = 0.25, outlier.colour = NA)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#74C476","#006D2C"), labels = c(agegroup2="2-5 yr",agegroup3="6+ yr"))+
scale_x_discrete(labels = c("Autumn"="Autumn (2016)","Winter"="Winter (2017)"))+
scale_y_continuous(transform = "log10")+
labs(x = "Season (year)", y = "Abundance (log EGP)", fill = "Age group")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable, nrow = 1,
labeller = labeller(variable = c(StrongylidetypeeggEPG="Trichostrongylidae eggs")))
ggplot(subset(TS1617.abund.df, variable %in% c("ProtostrongylideLarverLPG")),
aes(Municipality, value))+
geom_boxplot(aes(fill = Municipality), alpha = 0.25, outlier.colour = NA)+
geom_point(aes(fill = Municipality), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
geom_signif(comparisons = list(c("Stjørdal","Meråker")), annotations = c("."),
tip_length = 0.02)+
scale_y_continuous(expand = expansion(mult = c(0, 0.15)))+
labs(x = "Municipality", y = "Abundance", fill = "", title = "Autumn vs Winter (adult females)")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "none",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ variable, nrow = 1)
Richness
Number of parasite groups, excluding hatched strongylide (as already counted in Strongylide). Note there are two cases of ProtostrongylideLarver detection where E. alces or V. alces could not be specifically identified, therefore using Protostrongylide detection instead.
# also excluding mites because so few anyway, and we don't know if they are soil/food or host parasites
# two cases of ProtostrongylideLarver detection do NOT have E or V identified BUT E not found in this dataset, so can use ProtostrongylideLarver instead (as it is either unknown or V)
# table(TS1617[,c("EalcesLarvae01","ValcesLarvae01","ProtostrongylideLarver01")])
TS1617$num.of.parasite.groups <- rowSums(TS1617[,c("Eimeria01", "Strongylidetypeegg01", "Moniezia01", "NematodirusNematodirellaEgg01",
"Trichuris01", "Capillaria01", "Strongyloides01",
"ProtostrongylideLarver01","DictyocaulusLarver01")], na.rm = T)Using Poisson distribution, as richness is count data.
## MAIN
TS1617.richP <- glm(
formula = num.of.parasite.groups ~ Season + Adult_age_category + MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
data = TS1617,
family = poisson(link = "log"),
na.action = na.omit
)summary(TS1617.richP)##
## Call:
## glm(formula = num.of.parasite.groups ~ Season + Adult_age_category +
## MunicipalityDensity_Mean2016_18 + Days.harvested.vs.lab,
## family = poisson(link = "log"), data = TS1617, na.action = na.omit)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.673520 0.420558 1.601 0.109
## SeasonWinter 0.105147 0.245363 0.429 0.668
## Adult_age_category.L -0.137635 0.167055 -0.824 0.410
## MunicipalityDensity_Mean2016_18 -0.805360 0.680837 -1.183 0.237
## Days.harvested.vs.lab -0.008092 0.028569 -0.283 0.777
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 31.846 on 61 degrees of freedom
## Residual deviance: 29.208 on 57 degrees of freedom
## AIC: 168.03
##
## Number of Fisher Scoring iterations: 5performance::check_overdispersion(TS1617.richP)## # Overdispersion test
##
## dispersion ratio = 0.472
## Pearson's Chi-Squared = 26.892
## p-value = 1## No overdispersion detected.plot(allEffects(TS1617.richP))
Fig. S2c
TS1617.plot.richA <-
ggplot(TS1617, aes(Season, num.of.parasite.groups))+
# geom_boxplot(aes(fill = Season), outlier.colour = NA, alpha = 0.25)+
geom_violin(aes(fill = Season), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Season), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c(Autumn="#E57200",Winter="#004F71"))+
scale_x_discrete(labels = c(Autumn="Autumn\n2016",Winter="Winter\n2017"))+
scale_y_continuous(limits = c(0, round(max(c(TA16.sub$num.of.parasite.groups, TS1617$num.of.parasite.groups), na.rm = T)+1)))+
labs(x = "Season", y = "No. of parasite groups", fill = "", title = "Autumn vs Winter\n(adult females)")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))TS1617.plot.richA
ggplot(TS1617, aes(Season, num.of.parasite.groups))+
geom_boxplot(aes(fill = Adult_age_category), alpha = 0.25, outlier.colour = NA)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c("#74C476","#006D2C"), labels = c(agegroup2="2-5 yr",agegroup3="6+ yr"))+
scale_x_discrete(labels = c("Autumn"="Autumn (2016)","Winter"="Winter (2017)"))+
labs(x = "Season (year)", y = "Richness", fill = "Age group")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))
ggplot(TS1617, aes(MunicipalityDensity_Mean2016_18, num.of.parasite.groups))+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
geom_point(fill = "gold", shape = 21, color = "black", size = 3)+
labs(x = "Moose density (mean 2016-2018)", y = "Richness", title = "")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))## `geom_smooth()` using formula = 'y ~ x'
ggplot(TS1617, aes(MunicipalityDensity_Mean2016_18, num.of.parasite.groups))+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
geom_point(aes(fill = Adult_age_category), shape = 21, color = "black", size = 3)+
scale_fill_manual(values = c("#74C476","#006D2C"), labels = c(agegroup2="2-5 yr",agegroup3="6+ yr"))+
labs(x = "Moose density (mean 2016-2018)", y = "Richness", fill = "Age group")+
theme_classic()+
theme(axis.text = element_text(color = "black", size = 12), legend.position = "top",
plot.title = element_text(size = 13, hjust = 0.5),
axis.title = element_text(size = 12), strip.text.x = element_text(size = 9.5),
axis.text.x = element_text(angle = 0))+
facet_wrap(~ Season + Adult_age_category, nrow = 1)## `geom_smooth()` using formula = 'y ~ x'
TS1617$Municipality2 <- factor(
TS1617$Municipality,
levels = unique(TS1617[order(TS1617$MunicipalityDensity_Mean2016_18),"Municipality"]))
TA1617.plot.richA <-
ggplot(TS1617, aes(MunicipalityDensity_Mean2016_18, num.of.parasite.groups))+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
# geom_point(fill = "gold", shape = 21, color = "black", size = 3)+
geom_point(aes(fill = Municipality2), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.height = 0.1, jitter.width = 0.05))+
scale_y_continuous(limits = c(0,5))+
scale_x_continuous(limits = c(0.2,0.8))+
labs(x = "Moose density (mean 2016-2018)", y = "No. of parasite groups", fill = "")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))
TA1617.plot.richB <-
ggplot(TS1617, aes(Municipality2, num.of.parasite.groups))+
geom_violin(aes(fill = Municipality2), alpha = 0.25, draw_quantiles = 0.5)+
geom_point(aes(fill = Municipality2), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_y_continuous(limits = c(0,5))+
labs(x = "Municipality", y = "No. of parasite groups", fill = "", title = "Richness by municipality (adult females)")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))plot_grid(TA1617.plot.richB, TA1617.plot.richA,
align = "hv", axis = "tblr",
nrow = 2, labels = c("a","b"))
Table 4
Stat tables
# main table
TS1617.richP.summary <- summary(TS1617.richP)
colnames(TS1617.richP.summary$coefficients) <- c("estimate","std.error","statistic","p.value")
TS1617.richP.summary <- data.frame(
resp_var="num.of.parasite.groups",
covariates=row.names(TS1617.richP.summary$coefficients),
as.data.frame(TS1617.richP.summary$coefficients),
p.adj=NA,
row.names = NULL
)
TS1617.prev$summary[which(TS1617.prev$summary$covariates != "(Intercept)" & TS1617.prev$summary$p.value < 0.1),]## resp_var covariates estimate std.error statistic
## 3 Strongylidetypeegg01 Adult_age_category.L -1.519379 0.8426482 -1.8031
## p.value p.adj n.pos
## 3 0.07137245 0.5976107 57TS1617.abund$summary[which(TS1617.abund$summary$covariates != "(Intercept)" & TS1617.abund$summary$p.value < 0.1),]## resp_var covariates estimate std.error
## 3 StrongylidetypeeggEPG Adult_age_category.L -0.3246267 0.1858473
## 14 TrichurisEPG MunicipalityDensity_Mean2016_18 -3.5752370 2.1179319
## statistic p.value p.adj n.pos
## 3 -1.746739 0.08068262 0.3560971 53
## 14 -1.688079 0.09139602 0.3560971 10TS1617.richP.summary[which(TS1617.richP.summary$covariates != "(Intercept)" & TS1617.richP.summary$p.value < 0.1),]## [1] resp_var covariates estimate std.error statistic p.value p.adj
## <0 rows> (or 0-length row.names)stats.tab2 <- rbind.data.frame(
TS1617.prev$summary[which(TS1617.prev$summary$resp_var %in% c("Strongylidetypeegg01")),],
TS1617.abund$summary[which(TS1617.abund$summary$resp_var %in% c("StrongylidetypeeggEPG")),],
data.frame(TS1617.richP.summary, n.pos=length(which(!is.na(TS1617$num.of.parasite.groups))))
)
stats.tab2$coeff_print <- paste(round(stats.tab2$estimate, digits = 3),"\u00B1", round(stats.tab2$std.error, digits = 3))
stats.tab2$pval_print <- ifelse(stats.tab2$p.value >= 0.001, round(stats.tab2$p.value, digits = 3), "<0.001")
stats.tab2$padj_print <- ifelse(stats.tab2$p.adj >= 0.001, round(stats.tab2$p.adj, digits = 3), "<0.001")
stats.tab2[,c("resp_var","n.pos","covariates","coeff_print","pval_print","padj_print")]## resp_var n.pos covariates coeff_print
## 1 Strongylidetypeegg01 57 (Intercept) 1.805 ± 2.71
## 2 Strongylidetypeegg01 57 SeasonWinter -3.202 ± 2.228
## 3 Strongylidetypeegg01 57 Adult_age_category.L -1.519 ± 0.843
## 4 Strongylidetypeegg01 57 MunicipalityDensity_Mean2016_18 -2.777 ± 4.071
## 5 Strongylidetypeegg01 57 Days.harvested.vs.lab 0.591 ± 0.424
## 6 StrongylidetypeeggEPG 53 (Intercept) 5.168 ± 0.486
## 7 StrongylidetypeeggEPG 53 SeasonWinter -0.439 ± 0.302
## 8 StrongylidetypeeggEPG 53 Adult_age_category.L -0.325 ± 0.186
## 9 StrongylidetypeeggEPG 53 MunicipalityDensity_Mean2016_18 -0.763 ± 0.736
## 10 StrongylidetypeeggEPG 53 Days.harvested.vs.lab -0.028 ± 0.033
## 11 num.of.parasite.groups 62 (Intercept) 0.674 ± 0.421
## 12 num.of.parasite.groups 62 SeasonWinter 0.105 ± 0.245
## 13 num.of.parasite.groups 62 Adult_age_category.L -0.138 ± 0.167
## 14 num.of.parasite.groups 62 MunicipalityDensity_Mean2016_18 -0.805 ± 0.681
## 15 num.of.parasite.groups 62 Days.harvested.vs.lab -0.008 ± 0.029
## pval_print padj_print
## 1 0.505 0.919
## 2 0.151 0.598
## 3 0.071 0.598
## 4 0.495 0.919
## 5 0.163 0.598
## 6 <0.001 <0.001
## 7 0.146 0.418
## 8 0.081 0.356
## 9 0.3 0.501
## 10 0.398 0.612
## 11 0.109 <NA>
## 12 0.668 <NA>
## 13 0.41 <NA>
## 14 0.237 <NA>
## 15 0.777 <NA>write.table(stats.tab2[,c("resp_var","n.pos","covariates","coeff_print","pval_print","padj_print")],
"figures/table_statistic_models_seasonal.txt",
row.names = F, quote = F, sep = "\t", fileEncoding = "latin1")Other material for paper
Table 1
Combine summary statistics into one table & format
ms.tab.summary <- rbind.data.frame(ms.tab1, ms.tab2)
ms.tab.summary$Sex <- gsub("Male", "Male", ms.tab.summary$Sex)
ms.tab.summary$Sex <- gsub("Female", "Female", ms.tab.summary$Sex)
ms.tab.summary$Adult_age_category <- gsub("agegroup1", "1 yr", ms.tab.summary$Adult_age_category)
ms.tab.summary$Adult_age_category <- gsub("agegroup2", "2-5 yr", ms.tab.summary$Adult_age_category)
ms.tab.summary$Adult_age_category <- gsub("agegroup3", "6+ yr", ms.tab.summary$Adult_age_category)
write.table(ms.tab.summary[,c("Season","Sex","Adult_age_category","Carcass_mass_category","count",
grep("Median",names(ms.tab.summary), value = T))],
"figures/table_sampling_summary.txt",
row.names = F, quote = F, sep = "\t", fileEncoding = "latin1")Table 2
Combine heatmaps into one figure.
prev.fig1.header.df <- data.frame(
Season="Autumn",
TA16.det.by.group.hm[!duplicated(TA16.det.by.group.hm[,c("Sex","Adult_age_category","num_inds.MM","num_inds.BA","group")]),c("Sex","Adult_age_category","num_inds.MM","num_inds.BA","group")]
)
prev.fig1.header.df$Adult_age_category <- ifelse(prev.fig1.header.df$Adult_age_category == "agegroup1","1 yr",
ifelse(prev.fig1.header.df$Adult_age_category == "agegroup2", "2-5 yr", "6+ yr"))
prev.fig1.header.df <- melt(prev.fig1.header.df,
id.vars = "group", measure.vars = c("Season","Sex","Adult_age_category","num_inds.MM","num_inds.BA"))## Warning: attributes are not identical across measure variables; they will be
## droppedprev.fig1.header.df$group <- factor(as.character(prev.fig1.header.df$group),
levels = c("agegroup1_Male", "agegroup2_Male", "agegroup3_Male",
"agegroup1_Female", "agegroup2_Female", "agegroup3_Female"))
prev.fig1.header.df$variable <- as.character(prev.fig1.header.df$variable)
prev.fig1.header.df[which(prev.fig1.header.df$variable == "Adult_age_category"),"variable"] <- "Age group"
prev.fig1.header.df[which(prev.fig1.header.df$variable == "num_inds.MM"),"variable"] <- "n (McMaster)"
prev.fig1.header.df[which(prev.fig1.header.df$variable == "num_inds.BA"),"variable"] <- "n (Baermann)"
prev.fig1.header.df$variable <- factor(prev.fig1.header.df$variable,
levels = rev(c("Season","Sex","Age group","n (McMaster)","n (Baermann)")))
prev.fig1.header <-
ggplot(prev.fig1.header.df, aes(group, variable))+
geom_tile(fill = "white", colour = "grey50", linewidth = 0.5)+
geom_text(aes(label=value))+
theme_bw()+
guides(fill = guide_colorbar(barwidth = 10, barheight = 1))+
theme(axis.title = element_blank(),
axis.text.y = ggtext::element_markdown(size = 12, colour = "black", hjust = 1),
axis.text.x = element_blank(),
panel.border = element_blank(),
axis.ticks = element_blank()
)prev.fig2.header.df <- data.frame(
Sex="Female",
TS1617.det.by.group.hm[!duplicated(TS1617.det.by.group.hm[,c("Season","Adult_age_category","num_inds.MM","num_inds.BA","group")]),c("Season","Adult_age_category","num_inds.MM","num_inds.BA","group")]
)
prev.fig2.header.df$Adult_age_category <- ifelse(prev.fig2.header.df$Adult_age_category == "agegroup1","1 yr",
ifelse(prev.fig2.header.df$Adult_age_category == "agegroup2", "2-5 yr", "6+ yr"))
prev.fig2.header.df <- melt(prev.fig2.header.df,
id.vars = "group", measure.vars = c("Season","Sex","Adult_age_category","num_inds.MM","num_inds.BA"))## Warning: attributes are not identical across measure variables; they will be
## droppedprev.fig2.header.df$variable <- as.character(prev.fig2.header.df$variable)
prev.fig2.header.df[which(prev.fig2.header.df$variable == "Adult_age_category"),"variable"] <- "Age group"
prev.fig2.header.df[which(prev.fig2.header.df$variable == "num_inds.MM"),"variable"] <- "n (McMaster)"
prev.fig2.header.df[which(prev.fig2.header.df$variable == "num_inds.BA"),"variable"] <- "n (Baermann)"
prev.fig2.header.df$variable <- factor(prev.fig2.header.df$variable,
levels = rev(c("Season","Sex","Age group","n (McMaster)","n (Baermann)")))
prev.fig2.header <-
ggplot(prev.fig2.header.df, aes(group, variable))+
geom_tile(fill = "white", colour = "grey50", linewidth = 0.5)+
geom_text(aes(label=value))+
theme_bw()+
guides(fill = guide_colorbar(barwidth = 10, barheight = 1))+
theme(axis.title = element_blank(),
axis.text.y = ggtext::element_markdown(size = 12, colour = "black", hjust = 1),
axis.text.x = element_blank(),
panel.border = element_blank(),
axis.ticks = element_blank()
)plot_grid(
prev.fig1.header,
prev.fig2.header+
theme(axis.text.y = element_blank()),
prev.fig1+
theme(axis.text.x = element_blank()),
prev.fig2+
theme(axis.text.x = element_blank()),
align = "hv", axis = "trbl",
nrow = 2, ncol = 2,
rel_heights = c(0.5,1,0.5,1),
rel_widths = c(1,1,0.47,0.47),
labels = c("a","b")
)
# and then use Inkscape to remove some of the white space?
# ggsave("figures/PrevalenceHeatmap_by_SeasonSexAge_250228.png",units = "in", dpi = 1200, width = 14, height = 7)
# ggtext doesn't render correctly with ggsave, use dev insteadpng("figures/Table2_Prevalence_by_SeasonSexAge.png", units = "in", res = 1200, width = 12, height = 11)
plot_grid(
prev.fig1.header,
prev.fig2.header+
theme(axis.text.y = element_blank()),
prev.fig1+
theme(axis.text.x = element_blank()),
prev.fig2+
theme(axis.text.x = element_blank()),
align = "hv", axis = "trbl",
nrow = 2, ncol = 2,
rel_heights = c(0.5,1,0.5,1),
rel_widths = c(1,1,0.47,0.47),
labels = c("a","b")
)
dev.off()## png
## 2Fig. S2
plot_grid(TA16.plot.richA, TA16.plot.richB, TS1617.plot.richA,
align = "hv", axis = "tblr",
nrow = 1, labels = c("a","b","c"))## `geom_smooth()` using formula = 'y ~ x'
ggsave("figures/FigS2_Richness_by_AgeMassCatSeason.png",units = "in", dpi = 1200, width = 10, height = 4)Fig. S4
Weak associations with sex
TA16.det.by.Sex <- dcast(
TA16.det.df,
Adult_age_category + Sex ~ variable,
value.var = "value",
fun.aggregate = sum, fill = 0, drop = T, na.rm = T
)
## Calculate no. of individuals based on samples for which test was performed
TA16.det.by.Sex$num_inds.MM <- sapply(seq(1,nrow(TA16.det.by.Sex)), function(i){
age <- TA16.det.by.Sex[i,"Adult_age_category"]
sex <- TA16.det.by.Sex[i,"Sex"]
length(which(TA16.sub$Adult_age_category == age & TA16.sub$Sex == sex & TA16.sub$McMaster_performed == 1))
})
TA16.det.by.Sex$num_inds.BA <- sapply(seq(1,nrow(TA16.det.by.Sex)), function(i){
age <- TA16.det.by.Sex[i,"Adult_age_category"]
sex <- TA16.det.by.Sex[i,"Sex"]
length(which(TA16.sub$Adult_age_category == age & TA16.sub$Sex == sex & TA16.sub$Baermann_performed == 1))
})
TA16.det.by.Sex.hm <- melt(TA16.det.by.Sex,
id.vars = c("Adult_age_category","Sex","num_inds.MM","num_inds.BA"),
measure.vars = c(mcmaster.parasite.groups.01,baermann.parasite.groups.01),
variable.name = "parasite", value.name = "n.positive")
TA16.det.by.Sex.hm$proportion <- ifelse(
TA16.det.by.Sex.hm$parasite %in% mcmaster.parasite.groups.01,
TA16.det.by.Sex.hm$n.positive / TA16.det.by.Sex.hm$num_inds.MM,
TA16.det.by.Sex.hm$n.positive / TA16.det.by.Sex.hm$num_inds.BA
)
# add 95%CI to prevalence
TA16.det.by.Sex.hm$CI95.lower <- sapply(seq(1,nrow(TA16.det.by.Sex.hm)), function(i){
p <- as.character(TA16.det.by.Sex.hm[i,"parasite"])
n.det <- TA16.det.by.Sex.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TA16.det.by.Sex.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TA16.det.by.Sex.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[1])
})
TA16.det.by.Sex.hm$CI95.upper <- sapply(seq(1,nrow(TA16.det.by.Sex.hm)), function(i){
p <- as.character(TA16.det.by.Sex.hm[i,"parasite"])
n.det <- TA16.det.by.Sex.hm[i,"n.positive"]
if(p %in% mcmaster.parasite.groups.01){
n.tot <- TA16.det.by.Sex.hm[i,"num_inds.MM"]
} else if(p %in% baermann.parasite.groups.01){
n.tot <- TA16.det.by.Sex.hm[i,"num_inds.BA"]
} else {
return(NA)
}
t <- binom.test(n.det, n.tot, conf.level = 0.95)
return(t$conf.int[2])
})
TA16.det.by.Sex.hm$CI95.lower.v2 <- ifelse(TA16.det.by.Sex.hm$proportion == 0,NA,TA16.det.by.Sex.hm$CI95.lower)
TA16.det.by.Sex.hm$CI95.upper.v2 <- ifelse(TA16.det.by.Sex.hm$proportion == 0,NA,TA16.det.by.Sex.hm$CI95.upper)
TA16.det.by.Sex.hm$variable <- factor(TA16.det.by.Sex.hm$parasite,
levels = levels(TA16.det.df$variable))TA16.plot.sexA <-
ggplot(
subset(TA16.det.by.Sex.hm, variable %in% c("ProtostrongylideLarver01","EalcesLarvae01","ValcesLarvae01") & Adult_age_category == "agegroup1"))+
geom_errorbar(aes(x = Sex, ymin = CI95.lower.v2, ymax = CI95.upper.v2),
width = 0.5, linewidth = 0.75, colour = "black")+
geom_point(aes(Sex, proportion, fill = Sex, size = num_inds.BA), shape = 21)+
scale_fill_manual(values = c(Male="#E41A1C",Female="#377EB8"), labels = c(Male="Male", Female="Female"), guide = "none")+
scale_size(limits = c(10,50), range = c(1,4))+
labs(x = "Sex", y = "Prevalence", size = "Total no.\nof samples", fill = "Sex")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = ggtext::element_markdown(size = 13),
legend.position = "right", legend.box = "horizontal",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable + Adult_age_category, nrow = 1,
labeller = labeller(variable = c(ProtostrongylideLarver01="Protostrongylidae", EalcesLarvae01="*E. alces*", ValcesLarvae01="*V. alces*"),
Adult_age_category = c(agegroup1 = "1 yr")))TA16.plot.sexB <-
ggplot(subset(TA16.abund.df, variable %in% c("StrongylidetypeeggEPG")),
aes(Sex, value))+
geom_boxplot(aes(fill = Sex), outlier.colour = NA, alpha = 0.25)+
geom_point(aes(fill = Sex), shape = 21, color = "black", size = 3,
position = position_jitterdodge(jitter.width = 0.2))+
scale_fill_manual(values = c(Male="#E41A1C",Female="#377EB8"), labels = c(Male="Male", Female="Female"))+
scale_y_continuous(transform = "log10", expand = expansion(mult = c(0, 0.1)))+
labs(x = "Sex", y = "Abundance (log EPG)")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ variable + Adult_age_category, nrow = 1,
labeller = labeller(variable = c(StrongylidetypeeggEPG="Trichostrongylidae"),
Adult_age_category = c(agegroup1="1 yr", agegroup2="2-5 yr", agegroup3="6+ yr")))plot_grid(TA16.plot.sexA, TA16.plot.sexB,
align = "hv", axis = "tblr", nrow = 2,
labels = c("a","b"))
ggsave("figures/FigS4_Abundance_by_Sex.png",units = "in", dpi = 1200, width = 9, height = 6.5)Fig. S6
Harvest date vs postage vs date received in the lab
# autumn
summary(dat[which(dat$Season == "Autumn"),"Days.harvested.vs.posted"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.000 2.000 2.000 3.269 4.000 8.000summary(dat[which(dat$Season == "Autumn"),"Days.harvested.vs.lab"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.000 3.000 5.000 6.177 8.000 17.000summary(dat[which(dat$Season == "Autumn"),"Days.posted.vs.lab"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.000 1.000 1.000 2.908 3.000 13.000# winter
summary(dat[which(dat$Season == "Winter"),"Days.harvested.vs.posted"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 6.00 7.00 8.00 8.05 9.00 12.00summary(dat[which(dat$Season == "Winter"),"Days.harvested.vs.lab"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 7.00 8.00 9.00 9.05 10.00 13.00summary(dat[which(dat$Season == "Winter"),"Days.posted.vs.lab"])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1 1 1 1 1 1dat.plot.daysA <-
ggplot(subset(dat, !is.na(StrongylidetypeeggEPG)), aes(Days.harvested.vs.lab, StrongylidetypeeggEPG))+
geom_point(aes(fill = Season), shape = 21, color = "black", size = 3)+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
scale_fill_manual(values = c(Autumn="#E57200",Winter="#004F71"))+
scale_y_continuous(expand = expansion(mult = c(0, 0.1)))+
labs(x = "Days in transit", y = "Trichostrongylidae\nabundance (EPG)")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ Season, nrow = 1)
dat.plot.daysB <-
ggplot(subset(dat, !is.na(ProtostrongylideLarverLPG)), aes(Days.harvested.vs.lab, ProtostrongylideLarverLPG))+
geom_point(aes(fill = Season), shape = 21, color = "black", size = 3)+
geom_smooth(method=lm , color="black", fill="grey70", se=TRUE)+
scale_fill_manual(values = c(Autumn="#E57200",Winter="#004F71"))+
scale_y_continuous(expand = expansion(mult = c(0, 0.1)))+
labs(x = "Days in transit", y = "Protostrongylidae\nabundance (LPG)")+
theme_classic()+
theme(plot.title = element_text(size = 13, hjust = 0.5),
axis.text = element_text(color = "black", size = 12),
axis.title = element_text(size = 13),
axis.text.x = element_text(angle = 0),
strip.text.x = element_text(size = 13),
legend.position = "none",
legend.text = element_text(color = "black", size = 12),
legend.title = element_text(size = 13))+
facet_wrap(~ Season, nrow = 1)plot_grid(
dat.plot.daysA, dat.plot.daysB,
align = "hv", axis = "tblr",
nrow = 2, labels = c("a","b")
)
ggsave("figures/FigS6_Abundance_by_DaysInTransit.png",units = "in", dpi = 1200, height = 7, width = 8)Table S2
AIC model selection results
aic.res <- rbind.data.frame(
TA16.prev.aic,
TA16.abund.aic,
data.frame(resp_var="num.of.parasite.groups",TA16.richP.aic)
)
names(aic.res) <- c("parasite_group","degrees_of_freedom","AIC")
aic.res$model <- gsub("\\$.*","", gsub(".*\\.r","M",row.names(aic.res)))
aic.res$model_formula <- sapply(row.names(aic.res), function(x){
if(grepl("r1",x)){
"~ Exact_age + Carcass_mass_kg + Sex + MunicipalityDensity_Mean2016_18 + Days.in.transit"
} else if(grepl("r2",x)){
"~ Exact_age + Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18 + Days.in.transit"
} else if(grepl("r3",x)){
"~ Exact_age + Carcass_mass_category + Sex + MunicipalityDensity_2016 + Days.in.transit"
} else if(grepl("r4",x)){
"~ Exact_age + Carcass_mass_category + Sex + Municipality + Days.in.transit"
} else if(grepl("r5",x)){
"~ Exact_age + Carcass_mass_category * Sex + MunicipalityDensity_Mean2016_18 + Days.in.transit"
} else if(grepl("r6",x)){
"~ Exact_age * Carcass_mass_category + Sex + MunicipalityDensity_Mean2016_18 + Days.in.transit"
} else {
NA
}
})
write.table(aic.res[,c("model","model_formula","parasite_group","degrees_of_freedom","AIC")],
"figures/supp_table_AIC_comparisons_autumn.txt",
row.names = F, quote = F, sep = "\t", fileEncoding = "latin1")Table S3
Autumn - Stats table for all other models
# all supp models (+ parasites not included in main models)
TA16.rich.supp <- rbind.data.frame(
data.frame(model="M2", format_rich_summary(TA16.richP.r2), n.pos=length(which(!is.na(TA16.sub$num.of.parasite.groups)))),
data.frame(model="M3", format_rich_summary(TA16.richP.r3), n.pos=length(which(!is.na(TA16.sub$num.of.parasite.groups)))),
data.frame(model="M4", format_rich_summary(TA16.richP.r4), n.pos=length(which(!is.na(TA16.sub$num.of.parasite.groups)))),
data.frame(model="M5", format_rich_summary(TA16.richP.r5), n.pos=length(which(!is.na(TA16.sub$num.of.parasite.groups)))),
data.frame(model="M6", format_rich_summary(TA16.richP.r6), n.pos=length(which(!is.na(TA16.sub$num.of.parasite.groups))))
)
TA16.prev.supp <- rbind.data.frame(
data.frame(model="M2", TA16.prev.r2$summary[which(!TA16.prev.r2$summary$resp_var %in% c("ProtostrongylideLarver01","EalcesLarvae01","ValcesLarvae01")),]),
data.frame(model="M1", TA16.prev.r1$summary),
data.frame(model="M3", TA16.prev.r3$summary),
data.frame(model="M4", TA16.prev.r4$summary),
data.frame(model="M4.posthoc", format_posthoc_summary(TA16.prev.r4.posthoc.Trichuris, "Trichuris01"), n.pos=""),
data.frame(model="M4.posthoc", format_posthoc_summary(TA16.prev.r4.posthoc.Valces, "ValcesLarvae01"), n.pos=""),
data.frame(model="M5", TA16.prev.r5$summary),
data.frame(model="M6", TA16.prev.r6$summary)
)
TA16.abund.supp <- rbind.data.frame(
data.frame(model="M1", TA16.abund.r1$summary),
data.frame(model="M2", TA16.abund.r2$summary),
data.frame(model="M3", TA16.abund.r3$summary),
data.frame(model="M4", TA16.abund.r4$summary), # include all models, as non-posthoc aren't in the main text table,
data.frame(model="M4.posthoc", TA16.abund.r4.posthoc.Strongylide.summary, n.pos=""),
data.frame(model="M4.posthoc", TA16.abund.r4.posthoc.Protostrongylide.summary, n.pos=""),
data.frame(model="M4.posthoc", format_posthoc_summary(TA16.abund.r4.posthoc.Trichuris, "TrichurisEPG"), n.pos=""),
data.frame(model="M5", TA16.abund.r5$summary),
data.frame(model="M6", TA16.abund.r6$summary)
)supp.stats.tab1 <- rbind.data.frame(
TA16.prev.supp,
TA16.abund.supp,
TA16.rich.supp
)
supp.stats.tab1$coeff_print <- paste(round(supp.stats.tab1$estimate, digits = 3),"\u00B1", round(supp.stats.tab1$std.error, digits = 3))
supp.stats.tab1$pval_print <- ifelse(supp.stats.tab1$p.value >= 0.001, round(supp.stats.tab1$p.value, digits = 3), "<0.001")
supp.stats.tab1$padj_print <- ifelse(supp.stats.tab1$p.adj >= 0.001, round(supp.stats.tab1$p.adj, digits = 3), "<0.001")
supp.stats.tab1$sig_print <- ifelse(supp.stats.tab1$p.adj > 0.1,"",
ifelse(supp.stats.tab1$p.adj > 0.05,".",
ifelse(supp.stats.tab1$p.adj > 0.01,"*",
ifelse(supp.stats.tab1$p.adj > 0.001,"**","***"))))
supp.stats.tab1[is.na(supp.stats.tab1$sig_print),"sig_print"] <- ifelse(supp.stats.tab1[is.na(supp.stats.tab1$sig_print),"p.value"] > 0.1,"",
ifelse(supp.stats.tab1[is.na(supp.stats.tab1$sig_print),"p.value"] > 0.05,".",
ifelse(supp.stats.tab1[is.na(supp.stats.tab1$sig_print),"p.value"] > 0.01,"*",
ifelse(supp.stats.tab1[is.na(supp.stats.tab1$sig_print),"p.value"] > 0.001,"**","***"))))
supp.stats.tab1[which(supp.stats.tab1$covariates == "(Intercept)"),"sig_print"] <- ""
# format for direct inclusion in supp
resp_var.2.name <- c("Trichostrongylidae spp. detection", "Nematodirinae egg detection",
"Trichuris egg detection", "Strongyloides egg detection",
"Protostrongylidae larvae detection", "E. alces larvae detection",
"V. alces larvae detection", "Nematodirinae egg abundance", "Trichuris egg abundance",
"Trichostrongylidae egg abundance",
"Protostrongylidae larvae abundance", "Richness (no. of parasite groups)"
)
names(resp_var.2.name) <- c("Strongylidetypeegg01", "NematodirusNematodirellaEgg01", "Trichuris01",
"Strongyloides01", "ProtostrongylideLarver01", "EalcesLarvae01",
"ValcesLarvae01", "NematodirussppEPG", "TrichurisEPG",
"StrongylidetypeeggEPG",
"ProtostrongylideLarverLPG", "num.of.parasite.groups"
)
supp.stats.tab1$parasite_group <- sapply(supp.stats.tab1$resp_var, function(x){
as.character(resp_var.2.name[x])
})
covar.2.name <- c("(Intercept)", "Exact age", "Carcass mass category", "Sex",
"Mean moose density", "Days in transit", "Carcass mass (cont.)",
"Moose density (2016)", "Municipality: Meråker - Malvik", "Municipality: Selbu - Malvik",
"Municipality: Stjørdal - Malvik", "Municipality: Tydal - Malvik", "Municipality: Meråker - Malvik",
"Municipality: Selbu - Malvik", "Municipality: Stjørdal - Malvik", "Municipality: Tydal - Malvik", "Municipality: Selbu - Meråker",
"Municipality: Stjørdal - Meråker", "Municipality: Tydal - Meråker", "Municipality: Stjørdal - Selbu",
"Municipality: Tydal - Selbu", "Municipality: Tydal - Stjørdal",
"Carcass mass category:Sex", "Carcass mass category:Exact age"
)
names(covar.2.name) <- c("(Intercept)", "Exact_age", "Carcass_mass_category.L", "SexFemale",
"MunicipalityDensity_Mean2016_18", "Days.harvested.vs.lab", "Carcass_mass_kg",
"MunicipalityDensity_2016", "MunicipalityMeråker", "MunicipalitySelbu",
"MunicipalityStjørdal", "MunicipalityTydal", "Meråker - Malvik",
"Selbu - Malvik", "Stjørdal - Malvik", "Tydal - Malvik", "Selbu - Meråker",
"Stjørdal - Meråker", "Tydal - Meråker", "Stjørdal - Selbu",
"Tydal - Selbu", "Tydal - Stjørdal",
"Carcass_mass_category.L:SexFemale", "Exact_age:Carcass_mass_category.L"
)
supp.stats.tab1$covariates_print <- sapply(supp.stats.tab1$covariates, function(x){
as.character(covar.2.name[x])
})
head(supp.stats.tab1[,c("parasite_group","n.pos","covariates_print","coeff_print","pval_print","padj_print","sig_print")])## parasite_group n.pos covariates_print
## 1 Trichostrongylidae spp. detection 115 (Intercept)
## 2 Trichostrongylidae spp. detection 115 Exact age
## 3 Trichostrongylidae spp. detection 115 Carcass mass category
## 4 Trichostrongylidae spp. detection 115 Sex
## 5 Trichostrongylidae spp. detection 115 Mean moose density
## 6 Trichostrongylidae spp. detection 115 Days in transit
## coeff_print pval_print padj_print sig_print
## 1 11.459 ± 1671.036 0.995 0.996
## 2 -0.088 ± 0.176 0.618 0.838
## 3 11.903 ± 2363.196 0.996 0.996
## 4 0.19 ± 1.611 0.906 0.996
## 5 -1.345 ± 6.292 0.831 0.996
## 6 0.311 ± 0.434 0.474 0.701tail(supp.stats.tab1[,c("parasite_group","n.pos","covariates_print","coeff_print","pval_print","padj_print","sig_print")])## parasite_group n.pos covariates_print
## 513 Richness (no. of parasite groups) 117 Exact age
## 514 Richness (no. of parasite groups) 117 Carcass mass category
## 515 Richness (no. of parasite groups) 117 Sex
## 516 Richness (no. of parasite groups) 117 Mean moose density
## 517 Richness (no. of parasite groups) 117 Days in transit
## 518 Richness (no. of parasite groups) 117 Carcass mass category:Exact age
## coeff_print pval_print padj_print sig_print
## 513 -0.078 ± 0.034 0.022 <NA> *
## 514 0.059 ± 0.138 0.669 <NA>
## 515 -0.109 ± 0.129 0.399 <NA>
## 516 -0.25 ± 0.525 0.633 <NA>
## 517 0.017 ± 0.014 0.23 <NA>
## 518 -0.01 ± 0.047 0.836 <NA>write.table(supp.stats.tab1[,c("parasite_group","model","n.pos","covariates_print","coeff_print","pval_print","padj_print","sig_print")],
"figures/supp_table_all_statistic_models_autumn.txt",
row.names = F, quote = F, sep = "\t", fileEncoding = "latin1")Table S4
Autumn vs winter - Stats table for all other models
# parasites not included in main model
TS1617.prev.supp <- rbind.data.frame(
data.frame(model="", TS1617.prev$summary[which(!TS1617.prev$summary$resp_var %in% c("Strongylidetypeegg01")),])
)
TS1617.abund.supp <- rbind.data.frame(
data.frame(model="", TS1617.abund$summary[which(!TS1617.abund$summary$resp_var %in% c("StrongylidetypeeggEPG")),])
)supp.stats.tab2 <- rbind.data.frame(
TS1617.prev.supp,
TS1617.abund.supp
)
supp.stats.tab2$coeff_print <- paste(round(supp.stats.tab2$estimate, digits = 3),"\u00B1", round(supp.stats.tab2$std.error, digits = 3))
supp.stats.tab2$pval_print <- ifelse(supp.stats.tab2$p.value >= 0.001, round(supp.stats.tab2$p.value, digits = 3), "<0.001")
supp.stats.tab2$padj_print <- ifelse(supp.stats.tab2$p.adj >= 0.001, round(supp.stats.tab2$p.adj, digits = 3), "<0.001")
supp.stats.tab2$sig_print <- ifelse(supp.stats.tab2$p.adj > 0.1,"",
ifelse(supp.stats.tab2$p.adj > 0.05,".",
ifelse(supp.stats.tab2$p.adj > 0.01,"*",
ifelse(supp.stats.tab2$p.adj > 0.001,"**","***"))))
supp.stats.tab2[is.na(supp.stats.tab2$sig_print),"sig_print"] <- ifelse(supp.stats.tab2[is.na(supp.stats.tab2$sig_print),"p.value"] > 0.1,"",
ifelse(supp.stats.tab2[is.na(supp.stats.tab2$sig_print),"p.value"] > 0.05,".",
ifelse(supp.stats.tab2[is.na(supp.stats.tab2$sig_print),"p.value"] > 0.01,"*",
ifelse(supp.stats.tab2[is.na(supp.stats.tab2$sig_print),"p.value"] > 0.001,"**","***"))))
supp.stats.tab2[which(supp.stats.tab2$covariates == "(Intercept)"),"sig_print"] <- ""
# format for direct inclusion in supp
supp.stats.tab2$parasite_group <- sapply(supp.stats.tab2$resp_var, function(x){
as.character(resp_var.2.name[x])
})
covar.2.name <- append(covar.2.name, c(SeasonWinter="Season",
Adult_age_category.L="Age category (2-5 vs 6+ years)",
MunicipalityTrondheim="Municipality: Trondheim - Malvik"))
supp.stats.tab2$covariates_print <- sapply(supp.stats.tab2$covariates, function(x){
as.character(covar.2.name[x])
})
head(supp.stats.tab2[,c("parasite_group","n.pos","covariates_print","coeff_print","pval_print","padj_print","sig_print")])## parasite_group n.pos covariates_print
## 6 Nematodirinae egg detection 6 (Intercept)
## 7 Nematodirinae egg detection 6 Season
## 8 Nematodirinae egg detection 6 Age category (2-5 vs 6+ years)
## 9 Nematodirinae egg detection 6 Mean moose density
## 10 Nematodirinae egg detection 6 Days in transit
## 11 Trichuris egg detection 10 (Intercept)
## coeff_print pval_print padj_print sig_print
## 6 -0.648 ± 2.336 0.781 0.993
## 7 2.618 ± 2.328 0.261 0.598
## 8 -0.934 ± 0.818 0.254 0.598
## 9 0.833 ± 2.955 0.778 0.993
## 10 -0.565 ± 0.435 0.194 0.598
## 11 -0.055 ± 1.375 0.968 0.994tail(supp.stats.tab2[,c("parasite_group","n.pos","covariates_print","coeff_print","pval_print","padj_print","sig_print")])## parasite_group n.pos covariates_print
## 151 Trichuris egg abundance 10 Days in transit
## 161 Protostrongylidae larvae abundance 5 (Intercept)
## 171 Protostrongylidae larvae abundance 5 Season
## 181 Protostrongylidae larvae abundance 5 Age category (2-5 vs 6+ years)
## 191 Protostrongylidae larvae abundance 5 Mean moose density
## 201 Protostrongylidae larvae abundance 5 Days in transit
## coeff_print pval_print padj_print sig_print
## 151 -0.027 ± 0.089 0.766 0.958
## 161 -9.913 ± 9733.101 0.999 0.999
## 171 -0.09 ± 0.923 0.922 0.999
## 181 -14.684 ± 13764.684 0.999 0.999
## 191 -3.55 ± 2.816 0.207 0.501
## 201 0.004 ± 0.098 0.964 0.999write.table(supp.stats.tab2[,c("parasite_group","model","n.pos","covariates_print","coeff_print","pval_print","padj_print","sig_print")],
"figures/supp_table_all_statistic_models_seasonal.txt",
row.names = F, quote = F, sep = "\t", fileEncoding = "latin1")