Differential responses to pesticides and parasite stress are linked to bumble bee life stage
Emily Runnion
2024-10-17
Pollen Consumption
Input data
df <- read_csv("pollen.dates.csv", col_types = cols(treatment = col_factor(levels = c("1","2", "3", "4")), pollen.start = col_date(format = "%m/%d/%Y"),start.time = col_time(format = "%H:%M:%S"),pollen.end = col_date(format = "%m/%d/%Y"),end.time = col_time(format = "%H:%M:%S"),colony.start = col_date(format = "%m/%d/%Y")))
df$colony <- as.factor(df$colony)Pollen Analysis
#https://stackoverflow.com/questions/75580470/emtrends-in-presence-of-a-quadratic-term-difference-in-quadratic-trends
##NEW TIME Variable
df$pollen.time <- as.numeric(df$pollen.time)
df$block <- as.factor(df$block)
LM.fit <- lmer(whole_dif~crithidia*fungicide*poly(pollen.time,2)+(1|colony) + block, data = df)
LM.fit2 <- lmer(whole_dif~(crithidia+fungicide+poly(pollen.time,2))^2+(1|colony) + block, data = df)
LM.fit3 <- lmer(whole_dif~(crithidia+fungicide+poly(pollen.time,2))^2+(1|colony), data = df)
LM.fit4 <- lmer(whole_dif~(crithidia*fungicide+poly(pollen.time,2))^2+(1|colony), data = df)
summary(LM.fit2)## Linear mixed model fit by REML ['lmerMod']
## Formula: whole_dif ~ (crithidia + fungicide + poly(pollen.time, 2))^2 +
## (1 | colony) + block
## Data: df
##
## REML criterion at convergence: -344.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.6514 -0.5244 -0.1313 0.4560 4.6582
##
## Random effects:
## Groups Name Variance Std.Dev.
## colony (Intercept) 0.02156 0.1468
## Residual 0.03095 0.1759
## Number of obs: 733, groups: colony, 35
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 0.42954 0.08771 4.897
## crithidiaTRUE -0.12779 0.07454 -1.714
## fungicideTRUE -0.06971 0.07161 -0.974
## poly(pollen.time, 2)1 2.94132 0.31792 9.252
## poly(pollen.time, 2)2 -2.75860 0.32161 -8.577
## block4 0.45781 0.10748 4.259
## block6 -0.15034 0.10713 -1.403
## block7 0.02902 0.10743 0.270
## block8 0.13906 0.10737 1.295
## block9 0.04396 0.10738 0.409
## block10 0.21771 0.11704 1.860
## block11 -0.09109 0.10723 -0.849
## block12 0.07204 0.10725 0.672
## crithidiaTRUE:fungicideTRUE 0.02148 0.10331 0.208
## crithidiaTRUE:poly(pollen.time, 2)1 -1.35371 0.35888 -3.772
## crithidiaTRUE:poly(pollen.time, 2)2 0.81627 0.36106 2.261
## fungicideTRUE:poly(pollen.time, 2)1 -0.10258 0.35990 -0.285
## fungicideTRUE:poly(pollen.time, 2)2 0.87969 0.36215 2.429anova(LM.fit2, LM.fit, test = "Chi")## Data: df
## Models:
## LM.fit2: whole_dif ~ (crithidia + fungicide + poly(pollen.time, 2))^2 + (1 | colony) + block
## LM.fit: whole_dif ~ crithidia * fungicide * poly(pollen.time, 2) + (1 | colony) + block
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## LM.fit2 20 -352.13 -260.19 196.06 -392.13
## LM.fit 22 -349.80 -248.67 196.90 -393.80 1.6763 2 0.4325anova(LM.fit2, LM.fit3, test = "Chi")## Data: df
## Models:
## LM.fit3: whole_dif ~ (crithidia + fungicide + poly(pollen.time, 2))^2 + (1 | colony)
## LM.fit2: whole_dif ~ (crithidia + fungicide + poly(pollen.time, 2))^2 + (1 | colony) + block
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## LM.fit3 12 -330.81 -275.64 177.41 -354.81
## LM.fit2 20 -352.13 -260.19 196.06 -392.13 37.318 8 1.006e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1AIC(LM.fit2, LM.fit3, LM.fit4)## df AIC
## LM.fit2 20 -304.8886
## LM.fit3 12 -313.1797
## LM.fit4 14 -313.3530Anova(LM.fit3)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: whole_dif
## Chisq Df Pr(>Chisq)
## crithidia 2.7397 1 0.09788 .
## fungicide 0.4755 1 0.49048
## poly(pollen.time, 2) 279.3323 2 < 2.2e-16 ***
## crithidia:fungicide 0.0649 1 0.79894
## crithidia:poly(pollen.time, 2) 20.0616 2 4.402e-05 ***
## fungicide:poly(pollen.time, 2) 5.7454 2 0.05654 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(LM.fit)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: whole_dif
## Chisq Df Pr(>Chisq)
## crithidia 5.0655 1 0.02441 *
## fungicide 1.3478 1 0.24566
## poly(pollen.time, 2) 279.5854 2 < 2.2e-16 ***
## block 43.4624 8 7.187e-07 ***
## crithidia:fungicide 0.0431 1 0.83563
## crithidia:poly(pollen.time, 2) 20.0698 2 4.384e-05 ***
## fungicide:poly(pollen.time, 2) 6.0573 2 0.04838 *
## crithidia:fungicide:poly(pollen.time, 2) 1.7097 2 0.42536
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(LM.fit4)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: whole_dif
## Chisq Df Pr(>Chisq)
## crithidia 2.7377 1 0.09800 .
## fungicide 0.4751 1 0.49064
## poly(pollen.time, 2) 279.2567 2 < 2.2e-16 ***
## crithidia:fungicide 0.0648 1 0.79901
## crithidia:poly(pollen.time, 2) 20.0562 2 4.414e-05 ***
## fungicide:poly(pollen.time, 2) 5.7438 2 0.05659 .
## crithidia:fungicide:poly(pollen.time, 2) 1.7906 2 0.40848
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#LM.fit3 fits betterPlot pollen data
par(mar = c(5, 5, 4, 2) + 0.1)
df$time2 <- df$pollen.time^2
# Calculate fitted values
time.quad2 <- lme(whole_dif ~ treatment * pollen.time + treatment * time2,
random = list(colony = pdDiag(~ pollen.time)), data = df)
fitted3 <- fitted(time.quad2, level = 0)
# Create a data frame with ordered values
a <- with(df,
data.frame(pollen.time, fitted3, treatment)[order(treatment, pollen.time), ])
# Set the font size multiplier and symbol size multiplier
font_multiplier <- 2
symbol_multiplier <- 2 # Adjust this for larger symbols
legend_cex <- 1.5
# Assuming you have standard error data, here's an example
se <- with(a, tapply(fitted3, treatment, function(x) sd(x)/sqrt(length(x))))
# Plot
with(a, {
# Plot data for treatment 1
plot(pollen.time[treatment==1], fitted3[treatment==1],
ylab = "Average Pollen Consumption (g)", xlab = "Time (days)",
col = "darkorchid4", type = "b", pch = 1,
cex.lab = font_multiplier, # Size of axis labels
cex.axis = font_multiplier, # Size of axis numbers
cex = symbol_multiplier) # Size of symbols
# Add standard error ribbon for treatment 1
polygon(c(pollen.time[treatment==1], rev(pollen.time[treatment==1])),
c(fitted3[treatment==1] + se[1], rev(fitted3[treatment==1] - se[1])),
col = adjustcolor("darkorchid4", alpha.f = 0.2), border = NA)
# Add points for other treatments with increased symbol size
points(pollen.time[treatment==2], fitted3[treatment==2],
pch = 4, col = "dodgerblue", type = "b",
cex = symbol_multiplier) # Size of symbols
# Add standard error ribbon for treatment 2
polygon(c(pollen.time[treatment==2], rev(pollen.time[treatment==2])),
c(fitted3[treatment==2] + se[2], rev(fitted3[treatment==2] - se[2])),
col = adjustcolor("dodgerblue", alpha.f = 0.2), border = NA)
points(pollen.time[treatment==3], fitted3[treatment==3],
pch = 16, col = "orange4", type = "b",
cex = symbol_multiplier) # Size of symbols
# Add standard error ribbon for treatment 3
polygon(c(pollen.time[treatment==3], rev(pollen.time[treatment==3])),
c(fitted3[treatment==3] + se[3], rev(fitted3[treatment==3] - se[3])),
col = adjustcolor("orange4", alpha.f = 0.2), border = NA)
points(pollen.time[treatment==4], fitted3[treatment==4],
pch = 17, col = "orange", type = "b",
cex = symbol_multiplier) # Size of symbols
# Add standard error ribbon for treatment 4
polygon(c(pollen.time[treatment==4], rev(pollen.time[treatment==4])),
c(fitted3[treatment==4] + se[4], rev(fitted3[treatment==4] - se[4])),
col = adjustcolor("orange", alpha.f = 0.2), border = NA)
# Add a legend with larger font
legend("topleft", legend = c(" Control", "Fungicide", "Fungicide + Crithidia", "Crithidia"),
col = c("darkorchid4", "dodgerblue", "orange4", "orange"), pch = c(1, 4, 16, 17),
lty = 1, cex = legend_cex, bty= "n")
})
Disease dynamics
Input data
#### Data for qpCR Disease Dynamics
all_bees <- read_csv("qpcr_3.4_bees_all2.csv", col_types = cols(treatment = col_factor(levels = c("1",
"2", "3", "4")), replicate = col_factor(levels = c("1",
"4", "6", "7", "8", "9", "10", "11",
"12")), start = col_date(format = "%m/%d/%Y"),
innoculation_date = col_date(format = "%m/%d/%Y"),
date = col_date(format = "%m/%d/%Y")))## Warning: One or more parsing issues, call `problems()` on your data frame for details,
## e.g.:
## dat <- vroom(...)
## problems(dat)all_bees$colony <- as.factor(all_bees$colony)
all_bees$bee_id <- as.factor(all_bees$bee_id)
a_df <- read_csv("individual bees 3.4 average.csv",
col_types = cols(date_first_adl = col_date(format = "%m/%d/%Y"),
round = col_factor(levels = c("1",
"2", "3"))))
a_df$bee_id <- as.factor(a_df$bee_id)
a_df$colony <- as.factor(a_df$colony)
a_df$treatment <- as.factor(a_df$treatment)
a_df$replicate <- as.factor(a_df$replicate)
a_df_na <- na.omit(a_df)
qpcr <- read_csv("qpcr_3.4_bees_all.csv",
col_types = cols(inoculate_round = col_factor(levels = c("1",
"2", "3")), inoculate_01 = col_logical(),
`end date` = col_date(format = "%m/%d/%Y"),
treatment = col_factor(levels = c("3",
"4")), replicate = col_factor(levels = c("1",
"4", "5", "6", "7", "8", "9",
"10", "11", "12")), start = col_date(format = "%m/%d/%Y"),
innoculation_date = col_date(format = "%m/%d/%Y"),
date = col_date(format = "%m/%d/%Y"),
alive_or_dead = col_logical()))## Warning: One or more parsing issues, call `problems()` on your data frame for details,
## e.g.:
## dat <- vroom(...)
## problems(dat)qpcr$fungicide <- as.logical(qpcr$fungicide)
qpcr$crithidia <- as.logical(qpcr$crithidia)
qpcr$qro <- as.factor(qpcr$qro)
qpcr$colony <- as.factor(qpcr$colony)
qpcr$premature_death <- as.factor(qpcr$premature_death)
qpcr$dry <- as.double(qpcr$dry)
qpcr$bee_id <- as.factor(qpcr$bee_id)
qpcr$spore_include <- as.numeric(qpcr$spore_include)
qpcr$col_days <- as.factor(qpcr$col_days)
dfcox <- read_csv("data for cox ph.csv")
dfcox$bee_id <- as.factor(dfcox$bee_id)
dfcox$fungicide <- as.factor(dfcox$fungicide)
dfcox$crithidia <- as.factor(dfcox$crithidia)
dfcox$qro <- as.factor(dfcox$qro)
dfcox$inoculate_round <- as.factor(dfcox$inoculate_round)
dfcox$inoculate <- as.factor(dfcox$inoculate)
dfcox$inoculate_01 <- as.factor(dfcox$inoculate_01)
dfcox$premature_death <- as.factor(dfcox$premature_death)
dfcox$treatment <- as.factor(dfcox$treatment)
dfcox$replicate <- as.factor(dfcox$replicate)
dfcox$colony <- as.factor(dfcox$colony)
cbdf <- read_csv("cbdf.csv")
cbdf$colony <- as.factor(cbdf$colony)
cbdf$day <- as.factor(cbdf$day)
cbdf$fungicide <- as.factor(cbdf$fungicide)
cbdf$round <- as.factor(cbdf$round)
cbdf$block <- as.factor(cbdf$block)
qpcr_all <- read_csv("qpcr_3.4_bees_all_with5.csv")
qpcr_all.na <- na.omit(qpcr_all)
unique(qpcr_all$replicate)## [1] 1 4 5 6 7 8 9 10 11 12workers <- read_csv("workers.csv")
workers$colony <- as.factor(workers$colony)
workers$treatment <- as.factor(workers$treatment)
workers$block <- as.factor(workers$block)
workers$qro <- as.factor(workers$qro)
workers$inoculate <- as.logical(workers$inoculate)
custom_labels <- c("Control", "Fungicide", "Fungicide + Crithidia", "Crithidia")Worker survival probability analysis
#Over whole experiment
library(survival)
library(coxme)
library(survminer)
workers.na <- na.omit(workers)
workers.na$inoculate_round <- as.factor(workers.na$inoculate_round)
cox <- coxme(Surv(days_alive_since_inoc, inoc_censor) ~ crithidia + fungicide + inoculate_round + inoculate + (1|colony), data = workers.na)
Anova(cox)## Analysis of Deviance Table (Type II tests)
##
## Response: Surv(days_alive_since_inoc, inoc_censor)
## Df Chisq Pr(>Chisq)
## crithidia 1 4.5618 0.03269 *
## fungicide 1 0.5496 0.45846
## inoculate_round 2 6.5356 0.03809 *
## inoculate 1 3.7609 0.05246 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(cox)## Mixed effects coxme model
## Formula: Surv(days_alive_since_inoc, inoc_censor) ~ crithidia + fungicide + inoculate_round + inoculate + (1 | colony)
## Data: workers.na
##
## events, n = 53, 175
##
## Random effects:
## group variable sd variance
## 1 colony Intercept 0.8515921 0.7252091
## Chisq df p AIC BIC
## Integrated loglik 38.30 6.0 9.829e-07 26.30 14.47
## Penalized loglik 75.32 18.4 7.517e-09 38.53 2.28
##
## Fixed effects:
## coef exp(coef) se(coef) z p
## crithidia 0.9370 2.5523 0.4387 2.14 0.0327
## fungicide 0.3177 1.3740 0.4285 0.74 0.4585
## inoculate_round2 0.3573 1.4295 0.4951 0.72 0.4705
## inoculate_round3 1.4966 4.4665 0.5855 2.56 0.0106
## inoculateTRUE 0.6020 1.8258 0.3104 1.94 0.0525emmeans(cox, pairwise ~ crithidia*fungicide)## $emmeans
## crithidia fungicide emmean SE df asymp.LCL asymp.UCL
## 0 0 -0.110 0.359 Inf -0.813 0.594
## 1 0 0.827 0.336 Inf 0.170 1.485
## 0 1 0.208 0.349 Inf -0.476 0.892
## 1 1 1.145 0.347 Inf 0.465 1.825
##
## Results are averaged over the levels of: inoculate_round, inoculate
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95
##
## $contrasts
## contrast estimate SE df z.ratio
## crithidia0 fungicide0 - crithidia1 fungicide0 -0.937 0.439 Inf -2.136
## crithidia0 fungicide0 - crithidia0 fungicide1 -0.318 0.429 Inf -0.741
## crithidia0 fungicide0 - crithidia1 fungicide1 -1.255 0.625 Inf -2.006
## crithidia1 fungicide0 - crithidia0 fungicide1 0.619 0.601 Inf 1.031
## crithidia1 fungicide0 - crithidia1 fungicide1 -0.318 0.429 Inf -0.741
## crithidia0 fungicide1 - crithidia1 fungicide1 -0.937 0.439 Inf -2.136
## p.value
## 0.1418
## 0.8804
## 0.1856
## 0.7315
## 0.8804
## 0.1418
##
## Results are averaged over the levels of: inoculate_round, inoculate
## Results are given on the log (not the response) scale.
## P value adjustment: tukey method for comparing a family of 4 estimatesemmeans(cox, pairwise ~ inoculate)## $emmeans
## inoculate emmean SE df asymp.LCL asymp.UCL
## FALSE 0.217 0.152 Inf -0.081 0.514
## TRUE 0.819 0.281 Inf 0.268 1.369
##
## Results are averaged over the levels of: crithidia, fungicide, inoculate_round
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95
##
## $contrasts
## contrast estimate SE df z.ratio p.value
## FALSE - TRUE -0.602 0.31 Inf -1.939 0.0525
##
## Results are averaged over the levels of: crithidia, fungicide, inoculate_round
## Results are given on the log (not the response) scale.cox.t <- coxme(Surv(days_alive_since_inoc, inoc_censor) ~ treatment + inoculate_round + (1|colony), data = workers.na)
cox.emm <- emmeans(cox.t, pairwise ~ treatment)
cox.emm## $emmeans
## treatment emmean SE df asymp.LCL asymp.UCL
## 1 -0.715 0.468 Inf -1.6319 0.201
## 2 0.344 0.386 Inf -0.4125 1.101
## 3 0.796 0.361 Inf 0.0884 1.504
## 4 0.918 0.355 Inf 0.2219 1.613
##
## Results are averaged over the levels of: inoculate_round
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95
##
## $contrasts
## contrast estimate SE df z.ratio p.value
## treatment1 - treatment2 -1.060 0.683 Inf -1.551 0.4069
## treatment1 - treatment3 -1.512 0.657 Inf -2.300 0.0980
## treatment1 - treatment4 -1.633 0.664 Inf -2.459 0.0665
## treatment2 - treatment3 -0.452 0.552 Inf -0.819 0.8457
## treatment2 - treatment4 -0.573 0.553 Inf -1.036 0.7283
## treatment3 - treatment4 -0.121 0.522 Inf -0.232 0.9956
##
## Results are averaged over the levels of: inoculate_round
## Results are given on the log (not the response) scale.
## P value adjustment: tukey method for comparing a family of 4 estimatescox.inoc <- coxme(Surv(days_alive_since_inoc, inoc_censor) ~ treatment + inoculate_round + (1|colony), data = workers.na)
Anova(cox.inoc)## Analysis of Deviance Table (Type II tests)
##
## Response: Surv(days_alive_since_inoc, inoc_censor)
## Df Chisq Pr(>Chisq)
## treatment 3 6.9615 0.07313 .
## inoculate_round 2 6.7712 0.03386 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1cox.emm <- emmeans(cox.t, pairwise ~ treatment)
cox.emm## $emmeans
## treatment emmean SE df asymp.LCL asymp.UCL
## 1 -0.715 0.468 Inf -1.6319 0.201
## 2 0.344 0.386 Inf -0.4125 1.101
## 3 0.796 0.361 Inf 0.0884 1.504
## 4 0.918 0.355 Inf 0.2219 1.613
##
## Results are averaged over the levels of: inoculate_round
## Results are given on the log (not the response) scale.
## Confidence level used: 0.95
##
## $contrasts
## contrast estimate SE df z.ratio p.value
## treatment1 - treatment2 -1.060 0.683 Inf -1.551 0.4069
## treatment1 - treatment3 -1.512 0.657 Inf -2.300 0.0980
## treatment1 - treatment4 -1.633 0.664 Inf -2.459 0.0665
## treatment2 - treatment3 -0.452 0.552 Inf -0.819 0.8457
## treatment2 - treatment4 -0.573 0.553 Inf -1.036 0.7283
## treatment3 - treatment4 -0.121 0.522 Inf -0.232 0.9956
##
## Results are averaged over the levels of: inoculate_round
## Results are given on the log (not the response) scale.
## P value adjustment: tukey method for comparing a family of 4 estimatesbrood <- read_csv("brood.csv")
brood <- read_csv("brood.csv")
brood$colony <- as.factor(brood$colony)
brood$treatment <- as.factor(brood$treatment)
brood$block <- as.factor(brood$block)
brood$fungicide <- as.logical(brood$fungicide)
brood$crithidia <- as.logical(brood$crithidia)Plot COX PH Curves
ggsurvplot(
survfit(Surv(days_alive_since_inoc, inoc_censor) ~ treatment, data = workers.na),
legend.title = "",
censor.shape = 124,
censor.size = 2.5,
ylim = c(0.25, 1),
palette = c("darkorchid4", "dodgerblue", "orange4", "orange"),
legend.labs = c("Control", "Fungicide", "Fungicide + Crithidia", "Crithidia"),
font.x = 16,
font.y = 16,
font.tickslab = 16,
font.legend = 18,
legend = "right"
)
Likliehood of infection above detection limit
qpcr_all.na$adl_neg <- 1 - qpcr_all.na$adl
qpcr_all.na$round <- as.factor(qpcr_all.na$round)
qpcr_all.na$replicate <- as.factor(qpcr_all.na$replicate)
cbw2 <- glm(cbind(adl, adl_neg) ~ fungicide + round + inoculate + replicate, data = qpcr_all.na, family = binomial("logit"))
drop1(cbw2, test = "Chisq")## Single term deletions
##
## Model:
## cbind(adl, adl_neg) ~ fungicide + round + inoculate + replicate
## Df Deviance AIC LRT Pr(>Chi)
## <none> 1164.6 1192.6
## fungicide 1 1172.3 1198.3 7.656 0.0056575 **
## round 2 1178.6 1202.6 13.936 0.0009416 ***
## inoculate 1 1267.3 1293.3 102.632 < 2.2e-16 ***
## replicate 9 1224.1 1234.1 59.463 1.701e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(cbw2)## Analysis of Deviance Table (Type II tests)
##
## Response: cbind(adl, adl_neg)
## LR Chisq Df Pr(>Chisq)
## fungicide 7.656 1 0.0056575 **
## round 13.936 2 0.0009416 ***
## inoculate 102.632 1 < 2.2e-16 ***
## replicate 59.463 9 1.701e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(cbw2)##
## Call:
## glm(formula = cbind(adl, adl_neg) ~ fungicide + round + inoculate +
## replicate, family = binomial("logit"), data = qpcr_all.na)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.1059 0.2165 -9.725 < 2e-16 ***
## fungicideTRUE 0.4824 0.1748 2.759 0.00579 **
## round2 -1.6874 0.5422 -3.112 0.00186 **
## round3 1.0634 0.4727 2.250 0.02446 *
## inoculateTRUE 1.6621 0.1670 9.954 < 2e-16 ***
## replicate4 -0.4410 0.3231 -1.365 0.17225
## replicate5 2.8644 0.6075 4.715 2.41e-06 ***
## replicate6 0.2876 0.3651 0.788 0.43097
## replicate7 0.3445 0.5533 0.623 0.53356
## replicate8 2.4619 0.6048 4.071 4.68e-05 ***
## replicate9 -0.4547 0.3296 -1.380 0.16771
## replicate10 0.8307 0.3464 2.398 0.01649 *
## replicate11 1.2705 0.6384 1.990 0.04660 *
## replicate12 0.7912 0.3562 2.221 0.02635 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1354.0 on 1181 degrees of freedom
## Residual deviance: 1164.6 on 1168 degrees of freedom
## AIC: 1192.6
##
## Number of Fisher Scoring iterations: 4emm1 <- emmeans(cbw2, pairwise ~ fungicide, type = "response")
pairs(emm1)## contrast odds.ratio SE df null z.ratio p.value
## FALSE / TRUE 0.617 0.108 Inf 1 -2.759 0.0058
##
## Results are averaged over the levels of: round, inoculate, replicate
## Tests are performed on the log odds ratio scaleemm1## $emmeans
## fungicide prob SE df asymp.LCL asymp.UCL
## FALSE 0.335 0.0298 Inf 0.279 0.395
## TRUE 0.449 0.0405 Inf 0.371 0.529
##
## Results are averaged over the levels of: round, inoculate, replicate
## Confidence level used: 0.95
## Intervals are back-transformed from the logit scale
##
## $contrasts
## contrast odds.ratio SE df null z.ratio p.value
## FALSE / TRUE 0.617 0.108 Inf 1 -2.759 0.0058
##
## Results are averaged over the levels of: round, inoculate, replicate
## Tests are performed on the log odds ratio scaleem.df <- as.data.frame(emm1$emmeans)
em.df## fungicide prob SE df asymp.LCL asymp.UCL
## FALSE 0.3346131 0.02977979 Inf 0.2789763 0.3952630
## TRUE 0.4489313 0.04053016 Inf 0.3714293 0.5289951
##
## Results are averaged over the levels of: round, inoculate, replicate
## Confidence level used: 0.95
## Intervals are back-transformed from the logit scaleemm2 <- emmeans(cbw2, pairwise ~ inoculate, type = "response")
pairs(emm2)## contrast odds.ratio SE df null z.ratio p.value
## FALSE / TRUE 0.19 0.0317 Inf 1 -9.954 <.0001
##
## Results are averaged over the levels of: fungicide, round, replicate
## Tests are performed on the log odds ratio scaleemm2## $emmeans
## inoculate prob SE df asymp.LCL asymp.UCL
## FALSE 0.218 0.0209 Inf 0.180 0.262
## TRUE 0.595 0.0405 Inf 0.514 0.671
##
## Results are averaged over the levels of: fungicide, round, replicate
## Confidence level used: 0.95
## Intervals are back-transformed from the logit scale
##
## $contrasts
## contrast odds.ratio SE df null z.ratio p.value
## FALSE / TRUE 0.19 0.0317 Inf 1 -9.954 <.0001
##
## Results are averaged over the levels of: fungicide, round, replicate
## Tests are performed on the log odds ratio scaleVisulize likelihood of infection probabilities
ggplot(data = em.df, aes(x = fungicide, y = prob, fill = fungicide)) +
geom_col() +
geom_bar(stat = "identity", color = "black") +
geom_col_pattern(
aes(pattern = fungicide),
pattern_density = c(0, 0.4),
pattern_spacing = 0.03,
position = position_dodge(0.9)
) +
labs(x = NULL, y = "Probability of Testing Above Detection Limit") +
theme_cowplot() +
scale_x_discrete(labels = c("FALSE" = "Crithidia", "TRUE" = "Fungicide + Crithidia")) +
scale_fill_manual(values = c("FALSE" = "orange", "TRUE" = "orange4")) +
scale_pattern_manual(values = c("none", "stripe")) + # Customize bar colors
theme(legend.position = "none") + # Remove the legend
annotate(
geom = "text",
x = 0.8,
y = 0.45,
label = "P < 0.01",
size = 8
) +
geom_errorbar(aes(ymin = prob - SE, ymax = prob + SE), width = 0.2, size = 0.8, position = position_dodge(1)) +
theme(legend.position = "none",
axis.text = element_text(size = 20), # Set axis label font size
axis.title = element_text(size = 20)) + # Set axis title font size
theme(text = element_text(size = 16))## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
Days until first infection above detection limit and total times above detection limit
# Input data
adl <- read.csv("adl.csv")
adl$colony <- as.factor(adl$colony)
adl$treatment <-as.factor(adl$treatment)
adl$round <- as.factor(adl$round)
adl$inoculate <- as.logical(adl$inoculate)
adl$block <- as.factor(adl$block)
adlmod <- glmer.nb(days_to_adl ~ treatment + round + inoculate + block + (1|colony), data = adl)
drop1(adlmod, test = "Chisq")## Single term deletions
##
## Model:
## days_to_adl ~ treatment + round + inoculate + block + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 404.34
## treatment 1 403.34 0.9991 0.3175325
## round 2 415.07 14.7353 0.0006314 ***
## inoculate 1 410.09 7.7495 0.0053728 **
## block 9 406.33 19.9943 0.0179474 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(adlmod)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: days_to_adl
## Chisq Df Pr(>Chisq)
## treatment 0.9977 1 0.317870
## round 12.4901 2 0.001940 **
## inoculate 7.3987 1 0.006527 **
## block 20.9704 9 0.012782 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1adlemm <- emmeans(adlmod, pairwise ~ treatment, type = "response")
pairs(adlemm)## contrast ratio SE df null z.ratio p.value
## treatment3 / treatment4 0.866 0.124 Inf 1 -0.999 0.3179
##
## Results are averaged over the levels of: round, inoculate, block
## Tests are performed on the log scaleadlemm## $emmeans
## treatment response SE df asymp.LCL asymp.UCL
## 3 2.12 0.397 Inf 1.47 3.06
## 4 2.45 0.397 Inf 1.78 3.36
##
## Results are averaged over the levels of: round, inoculate, block
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast ratio SE df null z.ratio p.value
## treatment3 / treatment4 0.866 0.124 Inf 1 -0.999 0.3179
##
## Results are averaged over the levels of: round, inoculate, block
## Tests are performed on the log scaleadlemm <- emmeans(adlmod, pairwise ~ round, type = "response")
pairs(adlemm)## contrast ratio SE df null z.ratio p.value
## round1 / round2 1.85 0.745 Inf 1 1.530 0.2767
## round1 / round3 6.76 4.232 Inf 1 3.051 0.0065
## round2 / round3 3.65 2.816 Inf 1 1.678 0.2134
##
## Results are averaged over the levels of: treatment, inoculate, block
## P value adjustment: tukey method for comparing a family of 3 estimates
## Tests are performed on the log scaleadlemm## $emmeans
## round response SE df asymp.LCL asymp.UCL
## 1 5.288 0.936 Inf 3.738 7.48
## 2 2.857 0.846 Inf 1.598 5.10
## 3 0.783 0.457 Inf 0.249 2.46
##
## Results are averaged over the levels of: treatment, inoculate, block
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast ratio SE df null z.ratio p.value
## round1 / round2 1.85 0.745 Inf 1 1.530 0.2767
## round1 / round3 6.76 4.232 Inf 1 3.051 0.0065
## round2 / round3 3.65 2.816 Inf 1 1.678 0.2134
##
## Results are averaged over the levels of: treatment, inoculate, block
## P value adjustment: tukey method for comparing a family of 3 estimates
## Tests are performed on the log scaleadlemm <- emmeans(adlmod, pairwise ~ inoculate, type = "response")
pairs(adlemm)## contrast ratio SE df null z.ratio p.value
## FALSE / TRUE 1.62 0.289 Inf 1 2.720 0.0065
##
## Results are averaged over the levels of: treatment, round, block
## Tests are performed on the log scaleadlemm## $emmeans
## inoculate response SE df asymp.LCL asymp.UCL
## FALSE 2.90 0.449 Inf 2.14 3.93
## TRUE 1.79 0.370 Inf 1.19 2.68
##
## Results are averaged over the levels of: treatment, round, block
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast ratio SE df null z.ratio p.value
## FALSE / TRUE 1.62 0.289 Inf 1 2.720 0.0065
##
## Results are averaged over the levels of: treatment, round, block
## Tests are performed on the log scaledetadl <- glmer.nb(total.adl ~ treatment + inoculate + (1|colony), data = adl)
drop1(detadl, test = "Chisq")## Single term deletions
##
## Model:
## total.adl ~ treatment + inoculate + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 437.45
## treatment 1 435.98 0.5224 0.4698
## inoculate 1 466.19 30.7356 2.957e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(detadl)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: total.adl
## Chisq Df Pr(>Chisq)
## treatment 0.5418 1 0.4617
## inoculate 27.9557 1 1.241e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1adl.t.emm <- emmeans(detadl, pairwise ~ inoculate, type = "response")
pairs(adl.t.emm)## contrast ratio SE df null z.ratio p.value
## FALSE / TRUE 0.32 0.069 Inf 1 -5.287 <.0001
##
## Results are averaged over the levels of: treatment
## Tests are performed on the log scaleadl.t.emm## $emmeans
## inoculate response SE df asymp.LCL asymp.UCL
## FALSE 2.01 0.308 Inf 1.49 2.72
## TRUE 6.29 1.299 Inf 4.20 9.43
##
## Results are averaged over the levels of: treatment
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast ratio SE df null z.ratio p.value
## FALSE / TRUE 0.32 0.069 Inf 1 -5.287 <.0001
##
## Results are averaged over the levels of: treatment
## Tests are performed on the log scaleadl_sum <- adl %>%
group_by(treatment) %>%
summarise(wrkrs = length(bee_id))
adl_sum## # A tibble: 2 × 2
## treatment wrkrs
## <fct> <int>
## 1 3 50
## 2 4 49Maximum level of infection
adl$bee_id <- as.factor(adl$bee_id)
qpcr_max_sum <- qpcr %>%
group_by(bee_id, treatment, inoculate, inoculate_round, colony) %>%
summarise(max = max(spores_assumed))
max_sum <- qpcr_max_sum %>%
group_by(inoculate) %>%
summarise(mean = mean(max),
sd= sd(max),
n = length(max)) %>%
mutate(se = sd/sqrt(n))
hist(adl$max_infc)
adl$logmax <- log((adl$max_infc)+1)
hist(adl$logmax)
shapiro.test(adl$logmax)##
## Shapiro-Wilk normality test
##
## data: adl$logmax
## W = 0.90717, p-value = 3.447e-06range(adl$logmax)## [1] 0.000000 6.270988maxmod <- lmer(logmax ~ treatment + inoculate + round + (1|colony), data = adl)
Anova(maxmod)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: logmax
## Chisq Df Pr(>Chisq)
## treatment 0.2929 1 0.588367
## inoculate 9.8498 1 0.001698 **
## round 11.9265 2 0.002572 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1maxmod.emm <- emmeans(maxmod, pairwise ~ inoculate, type = "response")
library(NBZIMM)
max.zig = lme.zig(fixed = logmax ~ treatment + inoculate + round + offset(log(days_to_adl + 1)),
random = ~ 1 | colony, data = adl)## Computational iterations: 4
## Computational time: 0.001 minutesAnova(max.zig)## Analysis of Deviance Table (Type II tests)
##
## Response: zz
## Chisq Df Pr(>Chisq)
## treatment 2.3994 1 0.12138
## inoculate 7.9420 1 0.00483 **
## round 19.3761 2 6.202e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1max_sum <- adl %>%
group_by(treatment) %>%
summarise(m = mean(max_infc),
n = length(max_infc),
sd = sd(max_infc)) %>%
mutate(se = sd/sqrt(n))
max_sum## # A tibble: 2 × 5
## treatment m n sd se
## <fct> <dbl> <int> <dbl> <dbl>
## 1 3 99.7 50 124. 17.5
## 2 4 81.1 49 106. 15.1Infection level measured over time
# Input the data
qpcr <- read_csv("qpcr_time2.csv")
qpcr$bee_id <- as.factor(qpcr$bee_id)
qpcr$time_group <- as.factor(qpcr$time_group)
qpcr$inoculate_round <- as.factor(qpcr$inoculate_round)
qpcr$replicate <- as.factor(qpcr$replicate)
qpcr$fungicide <- as.factor(qpcr$fungicide)
# Look at shape
hist(qpcr$spores_assumed)
plot(qpcr$fungicide, qpcr$spores_assumed)
# Transform the data to normalize
qpcr$logspores <- log((qpcr$spores_assumed)+10)
shapiro.test(qpcr$logspores)##
## Shapiro-Wilk normality test
##
## data: qpcr$logspores
## W = 0.54946, p-value < 2.2e-16hist(qpcr$logspores)
range(qpcr$logspores)## [1] 2.302585 5.918894range(qpcr$spores_assumed)## [1] 0 362# Final selected model
f10 = glmer(logspores ~ time_group*fungicide + inoculate + inoculate_round + (1|bee_id), family = Gamma(link = "log"), data = qpcr)
Anova(f10)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: logspores
## Chisq Df Pr(>Chisq)
## time_group 4.1503 2 0.1255346
## fungicide 1.9378 1 0.1639072
## inoculate 13.5467 1 0.0002327 ***
## inoculate_round 31.4479 2 1.483e-07 ***
## time_group:fungicide 19.1116 2 7.079e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(f10)## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: Gamma ( log )
## Formula: logspores ~ time_group * fungicide + inoculate + inoculate_round +
## (1 | bee_id)
## Data: qpcr
##
## AIC BIC logLik deviance df.resid
## 1686.8 1741.6 -832.4 1664.8 1067
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.5267 -0.4995 -0.2416 0.0070 5.5692
##
## Random effects:
## Groups Name Variance Std.Dev.
## bee_id (Intercept) 0.00501 0.07078
## Residual 0.05020 0.22406
## Number of obs: 1078, groups: bee_id, 85
##
## Fixed effects:
## Estimate Std. Error t value Pr(>|z|)
## (Intercept) 0.93643 0.02634 35.548 < 2e-16 ***
## time_group2 -0.07806 0.02097 -3.722 0.000197 ***
## time_group3 -0.08184 0.02118 -3.864 0.000111 ***
## fungicideTRUE -0.03824 0.03226 -1.185 0.235854
## inoculateTRUE 0.12040 0.03271 3.681 0.000233 ***
## inoculate_round2 0.02122 0.03164 0.671 0.502515
## inoculate_round3 0.21184 0.03809 5.561 2.68e-08 ***
## time_group2:fungicideTRUE 0.12279 0.02981 4.120 3.79e-05 ***
## time_group3:fungicideTRUE 0.10416 0.03023 3.446 0.000570 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) tm_gr2 tm_gr3 fnTRUE inTRUE incl_2 incl_3 t_2:TR
## time_group2 -0.433
## time_group3 -0.421 0.535
## fungicdTRUE -0.606 0.345 0.336
## inocultTRUE -0.248 0.011 0.006 -0.006
## inoclt_rnd2 -0.322 -0.008 0.002 -0.070 -0.029
## inoclt_rnd3 -0.390 0.054 0.036 0.093 -0.010 0.247
## tm_gr2:TRUE 0.293 -0.702 -0.376 -0.482 -0.018 0.010 0.000
## tm_gr3:TRUE 0.280 -0.373 -0.699 -0.472 -0.009 0.010 0.015 0.537f10.emm <- emmeans(f10, pairwise ~ fungicide*time_group, type = "response")
f10.emm## $emmeans
## fungicide time_group response SE df asymp.LCL asymp.UCL
## FALSE 1 2.93 0.0737 Inf 2.79 3.08
## TRUE 1 2.82 0.0727 Inf 2.68 2.96
## FALSE 2 2.71 0.0678 Inf 2.58 2.84
## TRUE 2 2.95 0.0758 Inf 2.80 3.10
## FALSE 3 2.70 0.0680 Inf 2.57 2.83
## TRUE 3 2.88 0.0753 Inf 2.74 3.03
##
## Results are averaged over the levels of: inoculate, inoculate_round
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast ratio SE df null z.ratio p.value
## FALSE time_group1 / TRUE time_group1 1.039 0.0335 Inf 1 1.185 0.8440
## FALSE time_group1 / FALSE time_group2 1.081 0.0227 Inf 1 3.722 0.0027
## FALSE time_group1 / TRUE time_group2 0.994 0.0319 Inf 1 -0.202 1.0000
## FALSE time_group1 / FALSE time_group3 1.085 0.0230 Inf 1 3.864 0.0016
## FALSE time_group1 / TRUE time_group3 1.016 0.0329 Inf 1 0.492 0.9965
## TRUE time_group1 / FALSE time_group2 1.041 0.0331 Inf 1 1.250 0.8119
## TRUE time_group1 / TRUE time_group2 0.956 0.0203 Inf 1 -2.107 0.2836
## TRUE time_group1 / FALSE time_group3 1.045 0.0335 Inf 1 1.358 0.7520
## TRUE time_group1 / TRUE time_group3 0.978 0.0211 Inf 1 -1.033 0.9071
## FALSE time_group2 / TRUE time_group2 0.919 0.0291 Inf 1 -2.670 0.0813
## FALSE time_group2 / FALSE time_group3 1.004 0.0204 Inf 1 0.186 1.0000
## FALSE time_group2 / TRUE time_group3 0.940 0.0300 Inf 1 -1.947 0.3734
## TRUE time_group2 / FALSE time_group3 1.092 0.0349 Inf 1 2.766 0.0630
## TRUE time_group2 / TRUE time_group3 1.023 0.0210 Inf 1 1.091 0.8851
## FALSE time_group3 / TRUE time_group3 0.936 0.0301 Inf 1 -2.049 0.3147
##
## Results are averaged over the levels of: inoculate, inoculate_round
## P value adjustment: tukey method for comparing a family of 6 estimates
## Tests are performed on the log scaleworkcld10 <- cld(object = f10.emm,
adjust = "Tukey",
alpha = 0.05,
Letters = letters)
workcld10## fungicide time_group response SE df asymp.LCL asymp.UCL .group
## FALSE 3 2.70 0.0680 Inf 2.52 2.88 a
## FALSE 2 2.71 0.0678 Inf 2.54 2.89 a
## TRUE 1 2.82 0.0727 Inf 2.63 3.02 ab
## TRUE 3 2.88 0.0753 Inf 2.69 3.09 ab
## FALSE 1 2.93 0.0737 Inf 2.74 3.13 b
## TRUE 2 2.95 0.0758 Inf 2.75 3.15 ab
##
## Results are averaged over the levels of: inoculate, inoculate_round
## Confidence level used: 0.95
## Conf-level adjustment: sidak method for 6 estimates
## Intervals are back-transformed from the log scale
## P value adjustment: tukey method for comparing a family of 6 estimates
## Tests are performed on the log scale
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.f10.in <- emmeans(f10, pairwise ~ inoculate, type = "response")
f10.in$contrasts## contrast ratio SE df null z.ratio p.value
## FALSE / TRUE 0.887 0.029 Inf 1 -3.681 0.0002
##
## Results are averaged over the levels of: time_group, fungicide, inoculate_round
## Tests are performed on the log scalef10.in## $emmeans
## inoculate response SE df asymp.LCL asymp.UCL
## FALSE 2.66 0.0449 Inf 2.58 2.75
## TRUE 3.00 0.0897 Inf 2.83 3.19
##
## Results are averaged over the levels of: time_group, fungicide, inoculate_round
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast ratio SE df null z.ratio p.value
## FALSE / TRUE 0.887 0.029 Inf 1 -3.681 0.0002
##
## Results are averaged over the levels of: time_group, fungicide, inoculate_round
## Tests are performed on the log scaleworkcldin <- cld(object = f10.in,
adjust = "none",
alpha = 0.05,
Letters = letters)
workcldin## inoculate response SE df asymp.LCL asymp.UCL .group
## FALSE 2.66 0.0449 Inf 2.58 2.75 a
## TRUE 3.00 0.0897 Inf 2.83 3.19 b
##
## Results are averaged over the levels of: time_group, fungicide, inoculate_round
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## Tests are performed on the log scale
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.# Create various summaries based on different explanatory variables for later modeling
infec_sum1 <- qpcr %>%
group_by(fungicide, time_group, inoculate) %>%
summarise(m = mean(logspores),
n = length(logspores),
sd = sd(logspores)) %>%
mutate(se = sd/sqrt(n))
infec_sum1## # A tibble: 12 × 7
## # Groups: fungicide, time_group [6]
## fungicide time_group inoculate m n sd se
## <fct> <fct> <lgl> <dbl> <int> <dbl> <dbl>
## 1 FALSE 1 FALSE 2.74 133 0.875 0.0759
## 2 FALSE 1 TRUE 3.01 36 0.932 0.155
## 3 FALSE 2 FALSE 2.43 156 0.480 0.0384
## 4 FALSE 2 TRUE 2.92 38 0.739 0.120
## 5 FALSE 3 FALSE 2.45 147 0.458 0.0378
## 6 FALSE 3 TRUE 2.95 38 0.717 0.116
## 7 TRUE 1 FALSE 2.58 128 0.707 0.0625
## 8 TRUE 1 TRUE 2.74 35 0.869 0.147
## 9 TRUE 2 FALSE 2.62 147 0.707 0.0583
## 10 TRUE 2 TRUE 2.95 43 0.749 0.114
## 11 TRUE 3 FALSE 2.58 139 0.602 0.0511
## 12 TRUE 3 TRUE 2.82 38 0.742 0.120infec_sum2 <- qpcr %>%
group_by(fungicide, time_group) %>%
summarise(m = mean(spores_assumed),
n = length(spores_assumed),
sd = sd(spores_assumed)) %>%
mutate(se = sd/sqrt(n))
infec_sum2## # A tibble: 6 × 6
## # Groups: fungicide [2]
## fungicide time_group m n sd se
## <fct> <fct> <dbl> <int> <dbl> <dbl>
## 1 FALSE 1 18.8 169 46.3 3.56
## 2 FALSE 2 6.84 194 25.9 1.86
## 3 FALSE 3 6.36 185 19.0 1.40
## 4 TRUE 1 13.0 163 46.0 3.60
## 5 TRUE 2 13.2 190 40.3 2.92
## 6 TRUE 3 8.79 177 22.3 1.68infec_sum4 <- qpcr %>%
group_by(inoculate, time_group) %>%
summarise(m = mean(spores_assumed),
n = length(spores_assumed),
sd = sd(spores_assumed)) %>%
mutate(se = sd/sqrt(n))
infec_sum4## # A tibble: 6 × 6
## # Groups: inoculate [2]
## inoculate time_group m n sd se
## <lgl> <fct> <dbl> <int> <dbl> <dbl>
## 1 FALSE 1 14.3 261 44.3 2.74
## 2 FALSE 2 8.14 303 34.1 1.96
## 3 FALSE 3 5.70 286 19.2 1.13
## 4 TRUE 1 22.0 71 52.4 6.22
## 5 TRUE 2 16.8 81 32.5 3.61
## 6 TRUE 3 14.5 76 24.6 2.82infec_sum7 <- qpcr %>%
group_by(inoculate) %>%
summarise(m = mean(spores_assumed),
n = length(spores_assumed),
sd = sd(spores_assumed),
mean_maxinf = mean(max(spores_assumed))) %>%
mutate(se = sd/sqrt(n))
infec_sum7## # A tibble: 2 × 6
## inoculate m n sd mean_maxinf se
## <lgl> <dbl> <int> <dbl> <dbl> <dbl>
## 1 FALSE 9.23 850 33.9 362 1.16
## 2 TRUE 17.6 228 37.8 277 2.50infec_sum8 <- qpcr %>%
group_by(inoculate_round) %>%
summarise(m = mean(spores_assumed),
n = length(spores_assumed),
sd = sd(spores_assumed),
mean_maxinf = mean(max(spores_assumed))) %>%
mutate(se = sd/sqrt(n))
infec_sum8## # A tibble: 3 × 6
## inoculate_round m n sd mean_maxinf se
## <fct> <dbl> <int> <dbl> <dbl> <dbl>
## 1 1 6.99 657 26.5 355 1.03
## 2 2 10.6 275 34.4 325 2.08
## 3 3 29.9 146 56.7 362 4.69Plot infection levels per time period
# Combine workcld with infec_sum based on matching columns
infec_sum_3 <- merge(infec_sum2, workcld10[, c("fungicide", "time_group", ".group")],
by = c("fungicide", "time_group"), all.x = TRUE)
# Create the plot with annotations
ggplot(infec_sum_3, aes(x = time_group, y = m, fill = fungicide)) +
geom_col_pattern(
aes(pattern = fungicide),
pattern_spacing = 0.03,
position = position_dodge(0.9)
) +
coord_cartesian(ylim = c(0,28)) +
geom_errorbar(aes(ymin = m - se, ymax = m + se),
position = position_dodge(width = 0.9), width = 0.25) +
geom_text(aes(label = .group, y = m + se + 1),
position = position_dodge(width = 0.9),
size = 5,
hjust = .5) +
scale_fill_manual(values = c("orange", "orange4"),
labels = c("FALSE" = "Crithidia", "TRUE" = "Fungicide + Crithidia")) +
scale_pattern_manual(values = c("none", "stripe"), guide = "none") +
scale_x_discrete(
labels = c("1" = "Days 0-4", "2" = "Days 5-9", "3" = "Days 0-14")
) +
labs(fill = NULL,
y = "Cells/µl",
x = "") +
theme_cowplot()+
theme(
plot.title = element_text(size = 20, face = "bold"), # Title size
axis.title = element_text(size = 20), # Axis titles
axis.text = element_text(size = 20), # Axis text
legend.title = element_text(size = 20), # Legend title
legend.text = element_text(size = 20) # Legend text
) +
theme(legend.position = "right") +
annotate(
geom = "text",
x = 0.8,
y = 27,
label = "P < 0.01",
size = 8
)
Microcolony health metrics
Males
Input data
drones <- read_csv("drones.csv")
drones$treatment <- as.factor(drones$treatment)
drones$block <- as.factor(drones$block)
drones$colony <- as.factor(drones$colony)
drones$id <- as.factor(drones$id)
drones$abdomen_post_ee <- as.numeric(drones$abdomen_post_ee)
drones$fungicide <- as.factor(drones$fungicide)
drones$crithidia <- as.factor(drones$crithidia)
brood <- read_csv("brood.csv")Males Emerge Days
em.mod <- glmer.nb(emerge ~ fungicide + crithidia + dry_weight + workers_alive + block + (1|colony), data = drones)## Warning in theta.ml(Y, mu, weights = object@resp$weights, limit = limit, :
## iteration limit reachedsummary(em.mod)## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: Negative Binomial(10448843) ( log )
## Formula: emerge ~ fungicide + crithidia + dry_weight + workers_alive +
## block + (1 | colony)
## Data: drones
##
## AIC BIC logLik deviance df.resid
## 1605.0 1655.9 -788.5 1577.0 266
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.04860 -0.32634 -0.00853 0.30446 1.40901
##
## Random effects:
## Groups Name Variance Std.Dev.
## colony (Intercept) 4.351e-11 6.596e-06
## Number of obs: 280, groups: colony, 25
##
## Fixed effects:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 3.83790 0.06803 56.414 <2e-16 ***
## fungicideTRUE 0.02757 0.02192 1.258 0.2085
## crithidiaTRUE -0.01327 0.02227 -0.596 0.5513
## dry_weight -1.19294 0.08964 -13.308 <2e-16 ***
## workers_alive -0.02433 0.01268 -1.918 0.0551 .
## block4 -0.07057 0.04429 -1.593 0.1111
## block7 -0.05654 0.05191 -1.089 0.2760
## block8 -0.07821 0.05244 -1.492 0.1358
## block9 -0.05487 0.04895 -1.121 0.2623
## block10 -0.08649 0.04289 -2.017 0.0437 *
## block11 -0.05048 0.06536 -0.772 0.4400
## block12 -0.09124 0.05154 -1.770 0.0767 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) fnTRUE crTRUE dry_wg wrkrs_ block4 block7 block8 block9
## fungicdTRUE -0.395
## crithidTRUE -0.340 0.061
## dry_weight 0.031 -0.017 -0.036
## workers_alv -0.771 0.271 0.124 -0.098
## block4 -0.241 -0.064 0.072 0.037 -0.333
## block7 -0.443 0.179 0.129 0.017 -0.052 0.662
## block8 -0.288 -0.062 0.245 0.016 -0.240 0.725 0.593
## block9 -0.405 0.090 0.130 0.003 -0.093 0.687 0.611 0.616
## block10 -0.544 0.058 0.138 -0.014 0.032 0.729 0.667 0.660 0.676
## block11 -0.361 0.176 0.224 0.015 -0.087 0.563 0.522 0.523 0.519
## block12 -0.528 0.167 0.324 -0.015 0.019 0.646 0.617 0.618 0.616
## blck10 blck11
## fungicdTRUE
## crithidTRUE
## dry_weight
## workers_alv
## block4
## block7
## block8
## block9
## block10
## block11 0.554
## block12 0.680 0.543
## optimizer (Nelder_Mead) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')drop1(em.mod, test = "Chisq")## Single term deletions
##
## Model:
## emerge ~ fungicide + crithidia + dry_weight + workers_alive +
## block + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 1605.0
## fungicide 1 1604.5 1.4509 0.2284
## crithidia 1 1603.4 0.3331 0.5639
## dry_weight 1 1603.7 0.6426 0.4228
## workers_alive 1 1605.4 2.3678 0.1239
## block 7 1595.2 4.1697 0.7600em1 <- update(em.mod, .~. -block)
drop1(em1, test = "Chisq")## Single term deletions
##
## Model:
## emerge ~ fungicide + crithidia + dry_weight + workers_alive +
## (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 1595.2
## fungicide 1 1595.0 1.8500 0.17378
## crithidia 1 1593.2 0.0121 0.91234
## dry_weight 1 1595.1 1.9003 0.16805
## workers_alive 1 1596.5 3.2603 0.07097 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1em2 <- update(em1, .~. -dry_weight)
drop1(em2, test = "Chisq")## Single term deletions
##
## Model:
## emerge ~ fungicide + crithidia + workers_alive + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 1595.1
## fungicide 1 1595.0 1.8874 0.16950
## crithidia 1 1593.2 0.0613 0.80443
## workers_alive 1 1597.0 3.9384 0.04719 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(em2)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: emerge
## Chisq Df Pr(>Chisq)
## fungicide 2.1344 1 0.144024
## crithidia 0.0688 1 0.793044
## workers_alive 7.2682 1 0.007019 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(em2)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: emerge
## Chisq Df Pr(>Chisq)
## fungicide 2.1344 1 0.144024
## crithidia 0.0688 1 0.793044
## workers_alive 7.2682 1 0.007019 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(em2)## Generalized linear mixed model fit by maximum likelihood (Laplace
## Approximation) [glmerMod]
## Family: Negative Binomial(10448843) ( log )
## Formula: emerge ~ fungicide + crithidia + workers_alive + (1 | colony)
## Data: drones
##
## AIC BIC logLik deviance df.resid
## 1595.1 1616.9 -791.5 1583.1 274
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -0.88827 -0.34661 -0.06238 0.25737 1.88253
##
## Random effects:
## Groups Name Variance Std.Dev.
## colony (Intercept) 3.19e-10 1.786e-05
## Number of obs: 280, groups: colony, 25
##
## Fixed effects:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 3.717327 0.041214 90.195 < 2e-16 ***
## fungicideTRUE 0.027696 0.018958 1.461 0.14402
## crithidiaTRUE -0.005003 0.019069 -0.262 0.79304
## workers_alive -0.023682 0.008784 -2.696 0.00702 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) fnTRUE crTRUE
## fungicdTRUE -0.258
## crithidTRUE -0.223 -0.053
## workers_alv -0.938 0.082 0.058
## optimizer (Nelder_Mead) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')em_sum <- drones %>%
group_by(treatment) %>%
summarise(m = mean(emerge),
sd = sd(emerge),
n = length(emerge)) %>%
mutate(se = sd/sqrt(n))
em_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 37.0 2.41 102 0.238
## 2 2 38.1 2.53 68 0.307
## 3 3 38.4 4.08 49 0.583
## 4 4 36.7 2.54 61 0.325em_sum$plot <- em_sum$m + em_sum$seMales relative fat
hist(drones$relative_fat)
range(drones$relative_fat)## [1] -0.000150305 0.003399280shapiro.test(drones$relative_fat)##
## Shapiro-Wilk normality test
##
## data: drones$relative_fat
## W = 0.97859, p-value = 0.0003281drones$sqrt_rf <- (sqrt(drones$relative_fat))## Warning in sqrt(drones$relative_fat): NaNs produceddrones$logrf <- log(drones$relative_fat)## Warning in log(drones$relative_fat): NaNs producedshapiro.test(drones$logrf)##
## Shapiro-Wilk normality test
##
## data: drones$logrf
## W = 0.94405, p-value = 9.012e-09shapiro.test(drones$sqrt_rf)##
## Shapiro-Wilk normality test
##
## data: drones$sqrt_rf
## W = 0.99049, p-value = 0.06877rf.mod <- lmer(sqrt_rf ~ fungicide + crithidia + workers_alive + block + (1|colony), data = drones)
drop1(rf.mod, test = "Chisq")## Single term deletions
##
## Model:
## sqrt_rf ~ fungicide + crithidia + workers_alive + block + (1 |
## colony)
## npar AIC LRT Pr(Chi)
## <none> -2019.2
## fungicide 1 -2017.6 3.5558 0.0593382 .
## crithidia 1 -2017.8 3.3635 0.0666578 .
## workers_alive 1 -2021.2 0.0041 0.9486992
## block 7 -2007.6 25.5725 0.0006004 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1rf2 <- update(rf.mod, .~. -workers_alive)
drop1(rf2, test = "Chisq")## Single term deletions
##
## Model:
## sqrt_rf ~ fungicide + crithidia + block + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> -2021.2
## fungicide 1 -2019.2 4.0103 0.045222 *
## crithidia 1 -2019.6 3.5526 0.059451 .
## block 7 -2008.8 26.3561 0.000435 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(rf2)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: sqrt_rf
## Chisq Df Pr(>Chisq)
## fungicide 3.1661 1 0.0751799 .
## crithidia 2.7294 1 0.0985137 .
## block 26.4613 7 0.0004165 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1rf_sum <- drones %>%
group_by(treatment) %>%
summarise(m = mean(relative_fat),
sd = sd(relative_fat),
n = length(relative_fat)) %>%
mutate(se = sd/sqrt(n))
rf_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 0.00112 0.000482 102 0.0000477
## 2 2 0.00111 0.000408 68 0.0000495
## 3 3 0.00129 0.000508 49 0.0000725
## 4 4 0.00127 0.000435 61 0.0000557Males radial cell
hist(drones$radial_um)
shapiro.test(drones$radial_um)##
## Shapiro-Wilk normality test
##
## data: drones$radial_um
## W = 0.97721, p-value = 0.0001896range(drones$radial_um)## [1] 2073.526 3083.439drones$cuberc <- (drones$radial_um)^3
shapiro.test(drones$cuberc)##
## Shapiro-Wilk normality test
##
## data: drones$cuberc
## W = 0.99255, p-value = 0.1738hist(drones$cuberc)
rc_mod <- lmer(cuberc ~ fungicide + crithidia + block + workers_alive + (1|colony), data = drones)
drop1(rc_mod, test = "Chisq")## Single term deletions
##
## Model:
## cuberc ~ fungicide + crithidia + block + workers_alive + (1 |
## colony)
## npar AIC LRT Pr(Chi)
## <none> 13130
## fungicide 1 13136 7.6994 0.005524 **
## crithidia 1 13130 1.9668 0.160792
## block 7 13126 9.6578 0.208808
## workers_alive 1 13129 0.8686 0.351356
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1rc2 <- update(rc_mod, .~. -workers_alive)
drop1(rc2, test = "Chisq")## Single term deletions
##
## Model:
## cuberc ~ fungicide + crithidia + block + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 13129
## fungicide 1 13137 10.3573 0.00129 **
## crithidia 1 13128 1.4461 0.22916
## block 7 13126 11.0296 0.13733
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1rc3 <- update(rc2, .~. -block)
Anova(rc3)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: cuberc
## Chisq Df Pr(>Chisq)
## fungicide 4.9350 1 0.02632 *
## crithidia 0.3472 1 0.55568
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1rc_sum <- drones %>%
group_by(treatment) %>%
summarise(m = mean(radial_um),
sd = sd(radial_um),
n = length(radial_um)) %>%
mutate(se = sd/sqrt(n))
rc_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 2690. 188. 102 18.6
## 2 2 2664. 174. 68 21.1
## 3 3 2660. 173. 49 24.7
## 4 4 2749. 135. 61 17.3rc_sum$plot <- rc_sum$m + rc_sum$se
rc_sum## # A tibble: 4 × 6
## treatment m sd n se plot
## <fct> <dbl> <dbl> <int> <dbl> <dbl>
## 1 1 2690. 188. 102 18.6 2708.
## 2 2 2664. 174. 68 21.1 2685.
## 3 3 2660. 173. 49 24.7 2685.
## 4 4 2749. 135. 61 17.3 2766.rc_plot <- ggplot(rc_sum, aes(x = treatment, y = m, fill = treatment)) +
geom_bar(stat = "identity", color = "black") +
geom_col_pattern(
aes(pattern = treatment),
pattern_density = c(0, 0, 0.4, 0), # Add density for the fourth column
pattern_spacing = 0.03,
position = position_dodge(0.9)
) +
scale_fill_viridis_d() +
geom_errorbar(aes(ymin = m - se, ymax = m + se), width = 0.2, position = position_dodge(0.9)) +
labs(x = "Treatment", y = "Radial cell length (um)") +
theme_classic(base_size = 20) +
coord_cartesian(ylim=c(2600, 2800)) +
annotate(geom = "text",
x = 1, y = 2745,
label = "P = 0.02",
size = 8) +
theme(legend.position = "none",
axis.text = element_text(size = 20), # Set axis label font size
axis.title = element_text(size = 20)) + # Set axis title font size
theme(text = element_text(size = 20)) +
scale_fill_manual(values = c("darkorchid4", "dodgerblue", "orange4", "orange")) +
scale_pattern_manual(values = c("none", "none", "stripe", "none")) + # Add stripes to the fourth column
scale_x_discrete(labels = custom_labels) +
theme(legend.position = "none") +
geom_segment(x = 2, xend = 3, y = 2710, yend = 2710,
lineend = "round", linejoin = "round") +
geom_segment(x = 2, xend = 2, y = 2700, yend = 2720,
lineend = "round", linejoin = "round") +
geom_segment(x = 3, xend = 3, y = 2700, yend = 2720,
lineend = "round", linejoin = "round") +
geom_segment(x = 1, xend = 4, y = 2790, yend = 2790,
lineend = "round", linejoin = "round") +
geom_segment(x = 1, xend = 1, y = 2780, yend = 2800,
lineend = "round", linejoin = "round") +
geom_segment(x = 4, xend = 4, y = 2780, yend = 2800,
lineend = "round", linejoin = "round") +
geom_text(x = 2.5, y = 2712, label = "b", size = 6, vjust = -0.5) +
geom_text(x = 2.5, y = 2792, label = "a", size = 6, vjust = -0.5)
rc_plot
Total count of males
brood <- read_csv("brood.csv")
brood <- read_csv("brood.csv")
brood$colony <- as.factor(brood$colony)
brood$treatment <- as.factor(brood$treatment)
brood$block <- as.factor(brood$block)
brood$fungicide <- as.logical(brood$fungicide)
brood$crithidia <- as.logical(brood$crithidia)
male_count <- glm.nb(total_drones ~ fungicide + crithidia + block + workers_alive, data = brood)## Warning in glm.nb(total_drones ~ fungicide + crithidia + block + workers_alive,
## : alternation limit reacheddrop1(male_count, test = "Chisq")## Single term deletions
##
## Model:
## total_drones ~ fungicide + crithidia + block + workers_alive
## Df Deviance AIC LRT Pr(>Chi)
## <none> 42.183 190.34
## fungicide 1 42.673 188.83 0.490 0.4840
## crithidia 1 42.767 188.93 0.584 0.4446
## block 8 94.592 226.75 52.408 1.404e-08 ***
## workers_alive 1 72.919 219.08 30.736 2.957e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(male_count)## Analysis of Deviance Table (Type II tests)
##
## Response: total_drones
## LR Chisq Df Pr(>Chisq)
## fungicide 0.490 1 0.4840
## crithidia 0.584 1 0.4446
## block 52.408 8 1.404e-08 ***
## workers_alive 30.736 1 2.957e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(male_count)##
## Call:
## glm.nb(formula = total_drones ~ fungicide + crithidia + block +
## workers_alive, data = brood, init.theta = 6.826554202, link = log)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -6.452e-01 6.069e-01 -1.063 0.287785
## fungicideTRUE -1.500e-01 2.099e-01 -0.714 0.475035
## crithidiaTRUE -1.671e-01 2.126e-01 -0.786 0.431855
## block4 9.474e-01 4.131e-01 2.294 0.021808 *
## block6 -3.766e+01 3.355e+07 0.000 0.999999
## block7 5.753e-01 4.511e-01 1.275 0.202200
## block8 2.564e-01 4.383e-01 0.585 0.558634
## block9 1.972e-01 4.331e-01 0.455 0.648894
## block10 1.442e+00 4.027e-01 3.581 0.000342 ***
## block11 -2.076e-01 5.231e-01 -0.397 0.691532
## block12 8.826e-01 4.625e-01 1.908 0.056345 .
## workers_alive 5.700e-01 1.086e-01 5.247 1.55e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for Negative Binomial(6.8266) family taken to be 1)
##
## Null deviance: 172.432 on 35 degrees of freedom
## Residual deviance: 42.183 on 24 degrees of freedom
## AIC: 192.34
##
## Number of Fisher Scoring iterations: 1
##
##
## Theta: 6.83
## Std. Err.: 3.60
## Warning while fitting theta: alternation limit reached
##
## 2 x log-likelihood: -166.341m0 <- glmmTMB(total_drones ~ fungicide + crithidia + block + workers_alive, data = brood)
Anova(m0)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: total_drones
## Chisq Df Pr(>Chisq)
## fungicide 2.7355 1 0.09814 .
## crithidia 0.3672 1 0.54456
## block 58.4296 8 9.465e-10 ***
## workers_alive 18.9357 1 1.352e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1mc_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(total_drones),
n = length(total_drones),
sd = sd(total_drones)) %>%
mutate(se = sd/sqrt(n))
mc_sum## # A tibble: 4 × 5
## treatment m n sd se
## <fct> <dbl> <int> <dbl> <dbl>
## 1 1 11.3 9 7.84 2.61
## 2 2 7.67 9 7.37 2.46
## 3 3 5.44 9 8.16 2.72
## 4 4 6.89 9 7.18 2.39Male dry weight
hist(drones$dry_weight)
shapiro.test(drones$dry_weight)##
## Shapiro-Wilk normality test
##
## data: drones$dry_weight
## W = 0.99166, p-value = 0.1147range(drones$dry_weight)## [1] 0.0166 0.0541drones$log.dry_weight <- log((drones$dry_weight))
shapiro.test(drones$log.dry_weight)##
## Shapiro-Wilk normality test
##
## data: drones$log.dry_weight
## W = 0.95647, p-value = 2.014e-07hist(drones$log.dry_weight)
mdw <- lmer(dry_weight ~ fungicide + crithidia + block + workers_alive + (1|colony), data = drones )
drop1(mdw, test = "Chisq")## Single term deletions
##
## Model:
## dry_weight ~ fungicide + crithidia + block + workers_alive +
## (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> -1993.7
## fungicide 1 -1995.2 0.5266 0.468026
## crithidia 1 -1990.1 5.6389 0.017566 *
## block 7 -1987.5 20.2244 0.005105 **
## workers_alive 1 -1991.7 3.9608 0.046570 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(mdw)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: dry_weight
## Chisq Df Pr(>Chisq)
## fungicide 0.6628 1 0.41559
## crithidia 3.1104 1 0.07779 .
## block 17.0762 7 0.01691 *
## workers_alive 2.8255 1 0.09278 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1qqnorm(resid(mdw));qqline(resid(mdw))
male_dw <- drones %>%
group_by(treatment) %>%
summarise(m = mean(dry_weight),
sd = sd(dry_weight),
n = length(dry_weight)) %>%
mutate(se = sd/sqrt(n))
male_dw## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 0.0373 0.00752 102 0.000745
## 2 2 0.0376 0.00658 68 0.000798
## 3 3 0.0378 0.00706 49 0.00101
## 4 4 0.0393 0.00668 61 0.000855Worker dry weights
workers <- na.omit(workers)
workers <- read_csv("workers.csv")
workers$colony <- as.factor(workers$colony)
workers$treatment <- as.factor(workers$treatment)
workers$block <- as.factor(workers$block)
workers$qro <- as.factor(workers$qro)
workers$inoculate <- as.logical(workers$inoculate)workers$dry <- as.double(workers$dry)## Warning: NAs introduced by coercionhist(workers$dry)
shapiro.test(workers$dry)##
## Shapiro-Wilk normality test
##
## data: workers$dry
## W = 0.96135, p-value = 3.197e-05workers$logdry <- log(workers$dry)
shapiro.test(workers$logdry)##
## Shapiro-Wilk normality test
##
## data: workers$logdry
## W = 0.99094, p-value = 0.2537hist(workers$logdry)
workers$fungicide <- as.logical(workers$fungicide)
workers$crithidia <- as.logical(workers$crithidia)
wrkdry1 <- lmer(logdry ~ fungicide*crithidia + inoculate +block + (1|colony), data = workers)
drop1(wrkdry1, test = "Chisq")## Single term deletions
##
## Model:
## logdry ~ fungicide * crithidia + inoculate + block + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 93.353
## inoculate 1 91.391 0.0373 0.8468608
## block 9 104.116 28.7625 0.0007106 ***
## fungicide:crithidia 1 91.794 0.4412 0.5065339
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1wrkdry <- lmer(logdry ~ fungicide + crithidia + inoculate +block + (1|colony), data = workers)
drop1(wrkdry, test = "Chisq")## Single term deletions
##
## Model:
## logdry ~ fungicide + crithidia + inoculate + block + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 91.794
## fungicide 1 95.101 5.3065 0.0212462 *
## crithidia 1 102.738 12.9433 0.0003211 ***
## inoculate 1 89.833 0.0381 0.8452327
## block 9 102.295 28.5009 0.0007863 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1anova(wrkdry1, wrkdry, test = "Chisq")## Data: workers
## Models:
## wrkdry: logdry ~ fungicide + crithidia + inoculate + block + (1 | colony)
## wrkdry1: logdry ~ fungicide * crithidia + inoculate + block + (1 | colony)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## wrkdry 15 91.794 141.04 -30.897 61.794
## wrkdry1 16 93.353 145.88 -30.677 61.353 0.4412 1 0.5065Anova(wrkdry1)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: logdry
## Chisq Df Pr(>Chisq)
## fungicide 3.8656 1 0.0492852 *
## crithidia 10.4098 1 0.0012535 **
## inoculate 0.0413 1 0.8390495
## block 28.4252 9 0.0008097 ***
## fungicide:crithidia 0.2749 1 0.6000361
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(wrkdry)## Linear mixed model fit by REML ['lmerMod']
## Formula: logdry ~ fungicide + crithidia + inoculate + block + (1 | colony)
## Data: workers
##
## REML criterion at convergence: 104.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.91492 -0.56289 -0.00892 0.66798 2.04904
##
## Random effects:
## Groups Name Variance Std.Dev.
## colony (Intercept) 0.02217 0.1489
## Residual 0.07178 0.2679
## Number of obs: 197, groups: colony, 40
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) -2.781148 0.105194 -26.438
## fungicideTRUE -0.122302 0.061443 -1.990
## crithidiaTRUE -0.200682 0.061443 -3.266
## inoculateTRUE -0.009667 0.047465 -0.204
## block4 0.087618 0.135147 0.648
## block5 -0.156602 0.136922 -1.144
## block6 -0.008842 0.135923 -0.065
## block7 -0.100143 0.135922 -0.737
## block8 0.121194 0.135147 0.897
## block9 -0.386717 0.135147 -2.861
## block10 -0.126487 0.135147 -0.936
## block11 -0.378121 0.135922 -2.782
## block12 -0.156229 0.135147 -1.156wrkdry1 <- update(wrkdry, .~. -inoculate)
drop1(wrkdry1, test = "Chisq")## Single term deletions
##
## Model:
## logdry ~ fungicide + crithidia + block + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 89.833
## fungicide 1 93.139 5.3061 0.0212508 *
## crithidia 1 100.782 12.9492 0.0003201 ***
## block 9 100.343 28.5101 0.0007835 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1wm2 <- update(wrkdry1, .~. -fungicide)
drop1(wm2, test = "Chisq")## Single term deletions
##
## Model:
## logdry ~ crithidia + block + (1 | colony)
## npar AIC LRT Pr(Chi)
## <none> 93.139
## crithidia 1 102.998 11.859 0.0005738 ***
## block 9 100.838 25.700 0.0022873 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1anova(wrkdry, wm2, test = "Chisq")## Data: workers
## Models:
## wm2: logdry ~ crithidia + block + (1 | colony)
## wrkdry: logdry ~ fungicide + crithidia + inoculate + block + (1 | colony)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## wm2 13 93.139 135.82 -33.569 67.139
## wrkdry 15 91.794 141.04 -30.897 61.794 5.3442 2 0.06911 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(wrkdry1)## Linear mixed model fit by REML ['lmerMod']
## Formula: logdry ~ fungicide + crithidia + block + (1 | colony)
## Data: workers
##
## REML criterion at convergence: 100.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.91419 -0.57743 -0.00892 0.67578 2.02348
##
## Random effects:
## Groups Name Variance Std.Dev.
## colony (Intercept) 0.02225 0.1492
## Residual 0.07134 0.2671
## Number of obs: 197, groups: colony, 40
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) -2.78308 0.10475 -26.568
## fungicideTRUE -0.12228 0.06144 -1.990
## crithidiaTRUE -0.20072 0.06144 -3.267
## block4 0.08762 0.13513 0.648
## block5 -0.15659 0.13691 -1.144
## block6 -0.00897 0.13590 -0.066
## block7 -0.10021 0.13590 -0.737
## block8 0.12119 0.13513 0.897
## block9 -0.38672 0.13513 -2.862
## block10 -0.12649 0.13513 -0.936
## block11 -0.37822 0.13590 -2.783
## block12 -0.15623 0.13513 -1.156
##
## Correlation of Fixed Effects:
## (Intr) fnTRUE crTRUE block4 block5 block6 block7 block8 block9
## fungicdTRUE -0.286
## crithidTRUE -0.286 -0.023
## block4 -0.645 0.000 0.000
## block5 -0.637 0.115 -0.115 0.494
## block6 -0.641 -0.005 0.005 0.497 0.490
## block7 -0.638 -0.005 -0.005 0.497 0.491 0.494
## block8 -0.645 0.000 0.000 0.500 0.494 0.497 0.497
## block9 -0.645 0.000 0.000 0.500 0.494 0.497 0.497 0.500
## block10 -0.645 0.000 0.000 0.500 0.494 0.497 0.497 0.500 0.500
## block11 -0.644 0.005 0.005 0.497 0.491 0.494 0.494 0.497 0.497
## block12 -0.645 0.000 0.000 0.500 0.494 0.497 0.497 0.500 0.500
## blck10 blck11
## fungicdTRUE
## crithidTRUE
## block4
## block5
## block6
## block7
## block8
## block9
## block10
## block11 0.497
## block12 0.500 0.497Anova(wrkdry1)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: logdry
## Chisq Df Pr(>Chisq)
## fungicide 3.9617 1 0.0465471 *
## crithidia 10.6738 1 0.0010866 **
## block 29.1364 9 0.0006146 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1qqnorm(resid(wrkdry1));qqline(resid(wrkdry1))
pe <- emmeans(wrkdry1, pairwise ~ crithidia, type = "response")
pe## $emmeans
## crithidia emmean SE df lower.CL upper.CL
## FALSE -2.95 0.0443 27.9 -3.05 -2.86
## TRUE -3.16 0.0421 28.1 -3.24 -3.07
##
## Results are averaged over the levels of: fungicide, block
## Degrees-of-freedom method: kenward-roger
## Confidence level used: 0.95
##
## $contrasts
## contrast estimate SE df t.ratio p.value
## FALSE - TRUE 0.201 0.0614 28 3.267 0.0029
##
## Results are averaged over the levels of: fungicide, block
## Degrees-of-freedom method: kenward-rogerpairs(pe)## contrast estimate SE df t.ratio p.value
## FALSE - TRUE 0.201 0.0614 28 3.267 0.0029
##
## Results are averaged over the levels of: fungicide, block
## Degrees-of-freedom method: kenward-rogerpef <- emmeans(wrkdry1, pairwise ~ fungicide, type = "response")
pef## $emmeans
## fungicide emmean SE df lower.CL upper.CL
## FALSE -2.99 0.0421 28.1 -3.08 -2.91
## TRUE -3.12 0.0443 27.9 -3.21 -3.03
##
## Results are averaged over the levels of: crithidia, block
## Degrees-of-freedom method: kenward-roger
## Confidence level used: 0.95
##
## $contrasts
## contrast estimate SE df t.ratio p.value
## FALSE - TRUE 0.122 0.0614 28 1.990 0.0564
##
## Results are averaged over the levels of: crithidia, block
## Degrees-of-freedom method: kenward-rogerwrkdry.df <- na.omit(workers)
wrkdrysum <- wrkdry.df %>%
group_by(treatment) %>%
summarise(m = mean(dry),
sd = sd(dry),
n = length(dry)) %>%
mutate(se = sd/sqrt(n))
wrkdrysum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 0.0586 0.0186 44 0.00280
## 2 2 0.0528 0.0169 45 0.00252
## 3 3 0.0418 0.0128 42 0.00198
## 4 4 0.0488 0.0184 43 0.00281wtuk.means <- emmeans(object = wrkdry1,
specs = "crithidia",
adjust = "Tukey",
type = "response")
wtuk.means## crithidia emmean SE df lower.CL upper.CL
## FALSE -2.95 0.0443 27.9 -3.06 -2.85
## TRUE -3.16 0.0421 28.1 -3.25 -3.06
##
## Results are averaged over the levels of: fungicide, block
## Degrees-of-freedom method: kenward-roger
## Confidence level used: 0.95
## Conf-level adjustment: sidak method for 2 estimatesw.cld.model <- cld(object = wtuk.means,
adjust = "Tukey",
Letters = letters,
alpha = 0.05)
w.cld.model## crithidia emmean SE df lower.CL upper.CL .group
## TRUE -3.16 0.0421 28.1 -3.25 -3.06 a
## FALSE -2.95 0.0443 27.9 -3.06 -2.85 b
##
## Results are averaged over the levels of: fungicide, block
## Degrees-of-freedom method: kenward-roger
## Confidence level used: 0.95
## Conf-level adjustment: sidak method for 2 estimates
## significance level used: alpha = 0.05
## NOTE: If two or more means share the same grouping symbol,
## then we cannot show them to be different.
## But we also did not show them to be the same.blockdry <- emmeans(wrkdry1, pairwise ~ block, type = "response")
blockdrymeans <- as.data.frame(blockdry$emmeans)plot worker dry weights
work_dry <- ggplot(data = wrkdrysum, aes(x = treatment, y = m, fill = treatment)) +
geom_col(col = "black") +
coord_cartesian(ylim = c(0, 0.08)) +
labs(x = "Treatment", y = "Worker Dry Weight (g)") +
geom_col_pattern(
aes(pattern = treatment),
pattern_density = c(0, 0, 0.4, 0), # Add density for the fourth column
pattern_spacing = 0.03,
position = position_dodge(0.9)
) +
geom_errorbar(aes(ymin = m - se, ymax = m + se), width = 0.2, position = position_dodge(0.9)) +
theme_classic(base_size = 20) +
theme(legend.position = "none") +
annotate(
geom = "text",
x = 4,
y = 0.079,
label = "P < 0.01",
size = 7
) +
theme(legend.position = "none") +
scale_x_discrete(labels = custom_labels) +
scale_fill_manual(values = c("darkorchid4", "dodgerblue", "orange4", "orange")) +
scale_pattern_manual(values = c("none", "none", "stripe", "none")) +
geom_segment(x = 1, xend = 2, y = 0.07, yend = 0.07,
lineend = "round", linejoin = "round") +
geom_segment(x = 1, xend = 1, y = 0.07, yend = 0.069,
lineend = "round", linejoin = "round") +
geom_segment(x = 2, xend = 2, y = 0.07, yend = 0.069,
lineend = "round", linejoin = "round") +
geom_segment(x = 3, xend = 4, y = 0.06, yend = 0.06,
lineend = "round", linejoin = "round") +
geom_segment(x = 3, xend = 3, y = 0.06, yend = 0.059,
lineend = "round", linejoin = "round") +
geom_segment(x = 4, xend = 4, y = 0.06, yend = 0.059,
lineend = "round", linejoin = "round") +
geom_text(x = 1.5, y = 0.071, label = "a", size = 6, vjust = -0.5) +
geom_text(x = 3.5, y = 0.061, label = "b", size = 6, vjust = -0.5)
work_dry
Total brood cells
bcm <- glm.nb(brood_cells ~ fungicide + crithidia + workers_alive + block, data = brood)
Anova(bcm)## Analysis of Deviance Table (Type II tests)
##
## Response: brood_cells
## LR Chisq Df Pr(>Chisq)
## fungicide 3.070 1 0.07973 .
## crithidia 0.010 1 0.91928
## workers_alive 92.831 1 < 2e-16 ***
## block 102.709 8 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1b_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(brood_cells),
sd = sd(brood_cells),
n = length(brood_cells)) %>%
mutate(se = sd/sqrt(n))
b_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 30.7 23.1 9 7.71
## 2 2 29.8 21.1 9 7.04
## 3 3 21.7 24.9 9 8.30
## 4 4 23.9 26.3 9 8.77Honey pots
hpm <- glm(honey_pots ~ fungicide + crithidia + workers_alive + block, data = brood, family = "poisson")
Anova(hpm)## Analysis of Deviance Table (Type II tests)
##
## Response: honey_pots
## LR Chisq Df Pr(>Chisq)
## fungicide 3.5096 1 0.0610131 .
## crithidia 1.2041 1 0.2725029
## workers_alive 8.0172 1 0.0046336 **
## block 27.4941 8 0.0005806 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1hp_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(honey_pots),
sd = sd(honey_pots),
n = length(honey_pots)) %>%
mutate(se = sd/sqrt(n))
hp_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 3.33 2.5 9 0.833
## 2 2 4.11 3.02 9 1.01
## 3 3 2.56 2.24 9 0.747
## 4 4 1.89 1.90 9 0.633Larvae
tlm <- glm.nb(total_larvae ~ fungicide + crithidia + workers_alive + block, data = brood)
Anova(tlm)## Analysis of Deviance Table (Type II tests)
##
## Response: total_larvae
## LR Chisq Df Pr(>Chisq)
## fungicide 0.646 1 0.4217
## crithidia 0.037 1 0.8478
## workers_alive 39.129 1 3.967e-10 ***
## block 66.939 8 1.994e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1tl_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(total_larvae),
sd = sd(total_larvae),
n = length(total_larvae)) %>%
mutate(se = sd/sqrt(n))
tl_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 22.4 20.9 9 6.97
## 2 2 16.3 11.3 9 3.76
## 3 3 14.2 20.4 9 6.79
## 4 4 16 19.9 9 6.62dlm <- glm.nb(dead_larvae ~ fungicide + crithidia + workers_alive, data = brood)
Anova(dlm)## Analysis of Deviance Table (Type II tests)
##
## Response: dead_larvae
## LR Chisq Df Pr(>Chisq)
## fungicide 0.0725 1 0.78771
## crithidia 0.3865 1 0.53413
## workers_alive 4.4928 1 0.03404 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1dl_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(dead_larvae),
sd = sd(dead_larvae),
n = length(dead_larvae)) %>%
mutate(se = sd/sqrt(n))
dl_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 1.78 2.82 9 0.940
## 2 2 1.22 2.11 9 0.703
## 3 3 1.44 1.81 9 0.603
## 4 4 1.11 1.96 9 0.655llm <- glm.nb(live_larvae ~ fungicide + crithidia + workers_alive + block, data = brood)
Anova(llm)## Analysis of Deviance Table (Type II tests)
##
## Response: live_larvae
## LR Chisq Df Pr(>Chisq)
## fungicide 0.516 1 0.4723
## crithidia 0.002 1 0.9656
## workers_alive 44.398 1 2.680e-11 ***
## block 82.510 8 1.526e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1ll_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(live_larvae),
sd = sd(live_larvae),
n = length(live_larvae)) %>%
mutate(se = sd/sqrt(n))
ll_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 20.7 19.0 9 6.34
## 2 2 15.1 11.0 9 3.68
## 3 3 12.8 19.5 9 6.52
## 4 4 14.9 18.7 9 6.23plmod <- glm(cbind(live_larvae, dead_larvae) ~ fungicide + crithidia + block + workers_alive, data = brood, family = binomial("logit"))
Anova(plmod)## Analysis of Deviance Table (Type II tests)
##
## Response: cbind(live_larvae, dead_larvae)
## LR Chisq Df Pr(>Chisq)
## fungicide 0.0179 1 0.893715
## crithidia 0.0067 1 0.934615
## block 24.1726 8 0.002144 **
## workers_alive 6.1501 1 0.013140 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Pupae
tpm <- glm(total_pupae ~ fungicide + crithidia + workers_alive + block, data = brood, family = "poisson")
Anova(tpm)## Analysis of Deviance Table (Type II tests)
##
## Response: total_pupae
## LR Chisq Df Pr(>Chisq)
## fungicide 0.005 1 0.94611
## crithidia 4.885 1 0.02709 *
## workers_alive 51.869 1 5.932e-13 ***
## block 50.394 8 3.433e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1tp_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(total_pupae),
sd = sd(total_pupae),
n = length(total_pupae)) %>%
mutate(se = sd/sqrt(n))
tp_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 8.33 7.78 9 2.59
## 2 2 6 7.55 9 2.52
## 3 3 3.11 3.22 9 1.07
## 4 4 4.11 5.21 9 1.74dpm <- glm.nb(dead_pupae ~ fungicide + crithidia + workers_alive + block, data = brood)## Warning: glm.fit: fitted rates numerically 0 occurred## Warning in theta.ml(Y, mu, sum(w), w, limit = control$maxit, trace =
## control$trace > : iteration limit reached
## Warning in theta.ml(Y, mu, sum(w), w, limit = control$maxit, trace =
## control$trace > : iteration limit reachedAnova(dpm)## Analysis of Deviance Table (Type II tests)
##
## Response: dead_pupae
## LR Chisq Df Pr(>Chisq)
## fungicide 1.9743 1 0.15999
## crithidia 0.1183 1 0.73086
## workers_alive 4.5035 1 0.03383 *
## block 16.8182 8 0.03206 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1dp_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(dead_pupae),
sd = sd(dead_pupae),
n = length(dead_pupae)) %>%
mutate(se = sd/sqrt(n))
dp_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 0.111 0.333 9 0.111
## 2 2 0.111 0.333 9 0.111
## 3 3 0.222 0.441 9 0.147
## 4 4 0.111 0.333 9 0.111lpm <- glm(live_pupae ~ fungicide + crithidia + workers_alive + block, data = brood, family = "poisson")
Anova(lpm)## Analysis of Deviance Table (Type II tests)
##
## Response: live_pupae
## LR Chisq Df Pr(>Chisq)
## fungicide 0.002 1 0.96330
## crithidia 5.386 1 0.02029 *
## workers_alive 58.627 1 1.905e-14 ***
## block 45.223 8 3.339e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1lp_sum <- brood %>%
group_by(treatment) %>%
summarise(m = mean(live_pupae),
sd = sd(live_pupae),
n = length(live_pupae)) %>%
mutate(se = sd/sqrt(n))
lp_sum## # A tibble: 4 × 5
## treatment m sd n se
## <fct> <dbl> <dbl> <int> <dbl>
## 1 1 8.22 7.58 9 2.53
## 2 2 5.89 7.54 9 2.51
## 3 3 2.89 3.30 9 1.10
## 4 4 4 5.17 9 1.72ppmod <- glm(cbind(live_pupae, dead_pupae) ~ fungicide + crithidia + workers_alive, data = brood, family = binomial("logit"))
Anova(ppmod)## Analysis of Deviance Table (Type II tests)
##
## Response: cbind(live_pupae, dead_pupae)
## LR Chisq Df Pr(>Chisq)
## fungicide 0.0779 1 0.78011
## crithidia 0.0444 1 0.83313
## workers_alive 6.0550 1 0.01387 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1plot total pupae
live_pup <- ggplot(data = lp_sum, aes(x = treatment, y = m, fill = treatment)) +
geom_col(col = "black") +
coord_cartesian(ylim = c(0, 13)) +
labs(x = "Treatment", y = "Live Pupae Count") +
geom_col_pattern(
aes(pattern = treatment),
pattern_density = c(0, 0, 0.4, 0), # Add density for the fourth column
pattern_spacing = 0.03,
position = position_dodge(0.9)
) +
geom_errorbar(aes(ymin = m - se, ymax = m + se), width = 0.2, position = position_dodge(0.9)) +
theme_classic(base_size = 20) +
theme(legend.position = "none") +
annotate(
geom = "text",
x = 4,
y = 12,
label = "P = 0.02",
size = 7
) +
theme(legend.position = "none") +
scale_x_discrete(labels = custom_labels) +
scale_fill_manual(values = c("darkorchid4", "dodgerblue", "orange4", "orange")) +
scale_pattern_manual(values = c("none", "none", "stripe", "none")) +
geom_segment(x = 1, xend = 2, y = 12, yend = 12,
lineend = "round", linejoin = "round") +
geom_segment(x = 1, xend = 1, y = 11.7, yend = 12.3,
lineend = "round", linejoin = "round") +
geom_segment(x = 2, xend = 2, y = 11.7, yend = 12.3,
lineend = "round", linejoin = "round") +
geom_segment(x = 3, xend = 4, y = 7, yend = 7,
lineend = "round", linejoin = "round") +
geom_segment(x = 3, xend = 3, y = 6.7, yend = 7.3,
lineend = "round", linejoin = "round") +
geom_segment(x = 4, xend = 4, y = 6.7, yend = 7.3,
lineend = "round", linejoin = "round") +
geom_text(x = 1.5, y = 12.1, label = "a", size = 6, vjust = -0.5) +
geom_text(x = 3.5, y = 7.1, label = "b", size = 6, vjust = -0.5)
live_pup
tp_plot <- ggplot(data = tp_sum, aes(x = treatment, y = m, fill = treatment)) +
geom_col(col = "black") +
coord_cartesian(ylim = c(0, 13)) +
scale_fill_viridis_d() +
labs(x = "Treatment", y = "Total Pupae Count") +
geom_col_pattern(
aes(pattern = treatment),
pattern_density = c(0, 0, 0.4, 0), # Add density for the fourth column
pattern_spacing = 0.03,
position = position_dodge(0.9)
) +
geom_errorbar(aes(ymin = m - se, ymax = m + se), width = 0.2, position = position_dodge(0.9)) +
theme_classic(base_size = 20) +
theme(legend.position = "none") +
annotate(
geom = "text",
x = 4,
y = 12,
label = "P = 0.03",
size = 7
) +
theme(legend.position = "none") +
scale_x_discrete(labels = custom_labels) +
scale_fill_manual(values = c("darkorchid4", "dodgerblue", "orange4", "orange")) +
scale_pattern_manual(values = c("none", "none", "stripe", "none")) +
geom_segment(x = 1, xend = 2, y = 12, yend = 12,
lineend = "round", linejoin = "round") +
geom_segment(x = 1, xend = 1, y = 11.7, yend = 12.3,
lineend = "round", linejoin = "round") +
geom_segment(x = 2, xend = 2, y = 11.7, yend = 12.3,
lineend = "round", linejoin = "round") +
geom_segment(x = 3, xend = 4, y = 7, yend = 7,
lineend = "round", linejoin = "round") +
geom_segment(x = 3, xend = 3, y = 6.7, yend = 7.3,
lineend = "round", linejoin = "round") +
geom_segment(x = 4, xend = 4, y = 6.7, yend = 7.3,
lineend = "round", linejoin = "round") +
geom_text(x = 1.5, y = 12.1, label = "a", size = 6, vjust = -0.5) +
geom_text(x = 3.5, y = 7.1, label = "b", size = 6, vjust = -0.5)
tp_plot
plot_grid(tp_plot, live_pup, ncol=1, nrow =2)
Eggs
egg.mod <- glm.nb(eggs ~ fungicide + crithidia + workers_alive + block, data = brood)
drop1(egg.mod, test = "Chisq")## Single term deletions
##
## Model:
## eggs ~ fungicide + crithidia + workers_alive + block
## Df Deviance AIC LRT Pr(>Chi)
## <none> 46.045 281.46
## fungicide 1 46.605 280.02 0.5603 0.454128
## crithidia 1 46.053 279.46 0.0079 0.929219
## workers_alive 1 61.824 295.24 15.7787 7.12e-05 ***
## block 8 67.938 287.35 21.8931 0.005118 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Anova(egg.mod)## Analysis of Deviance Table (Type II tests)
##
## Response: eggs
## LR Chisq Df Pr(>Chisq)
## fungicide 0.5603 1 0.454128
## crithidia 0.0079 1 0.929219
## workers_alive 15.7787 1 7.12e-05 ***
## block 21.8931 8 0.005118 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Figures for supplemental material
Spatial Block summary
block_sum <- brood %>%
group_by(block) %>%
summarise(
mbc = mean(brood_cells),
sdbc = sd(brood_cells),
nbc = length(brood_cells),
# Larvae
ml = mean(live_larvae),
sdl = sd(live_larvae),
nl = length(live_larvae),
mdl = mean(dead_larvae),
sddl = sd(dead_larvae),
ndl = length(dead_larvae),
mtl = mean(total_larvae),
sdtl = sd(total_larvae),
ntl = length(total_larvae),
# Pupae
mp = mean(live_pupae),
sdp = sd(live_pupae),
np = length(live_pupae),
mdp = mean(dead_pupae),
sddp = sd(dead_pupae),
ndp = length(dead_pupae),
mtp = mean(total_pupae),
sdtp = sd(total_pupae),
ntp = length(total_pupae),
# Eggs
me = mean(eggs),
sde = sd(eggs),
ne = length(eggs),
# Honey pots
mhp = mean(honey_pots),
sdhp = sd(honey_pots),
nhp = length(honey_pots),
# Drones
mtd = mean(total_drones),
sdtd = sd(total_drones),
ntd = length(total_drones)
)
#write.csv(block_sum, file = "C:/Users/runni/OneDrive - The Ohio State University/Runnion and Sivakoff/Synergism Experiment/Disease Dynamics Manuscript/Data and code/block_sum.csv", row.names = FALSE)library(dplyr)
library(tidyr)
library(ggplot2)
block_sum <- brood %>%
group_by(block) %>%
summarise(
mbc = mean(brood_cells),
sdbc = sd(brood_cells),
nbc = length(brood_cells),
# Larvae
ml = mean(live_larvae),
sdl = sd(live_larvae),
nl = length(live_larvae),
mdl = mean(dead_larvae),
sddl = sd(dead_larvae),
ndl = length(dead_larvae),
mtl = mean(total_larvae),
sdtl = sd(total_larvae),
ntl = length(total_larvae),
# Pupae
mp = mean(live_pupae),
sdp = sd(live_pupae),
np = length(live_pupae),
mdp = mean(dead_pupae),
sddp = sd(dead_pupae),
ndp = length(dead_pupae),
mtp = mean(total_pupae),
sdtp = sd(total_pupae),
ntp = length(total_pupae),
# Eggs
me = mean(eggs),
sde = sd(eggs),
ne = length(eggs),
# Honey pots
mhp = mean(honey_pots),
sdhp = sd(honey_pots),
nhp = length(honey_pots),
# Drones
mtd = mean(total_drones),
sdtd = sd(total_drones),
ntd = length(total_drones)
)
block_sum_long <- block_sum %>%
pivot_longer(cols = starts_with("m"), # Select all mean columns
names_to = "variable",
values_to = "mean_value") %>%
mutate(variable = factor(variable,
levels = c("ml", "mdl", "mtl", "mp", "mdp", "mtp",
"me", "mhp", "mtd"),
labels = c("Live Larvae", "Dead Larvae", "Total Larvae",
"Live Pupae", "Dead Pupae", "Total Pupae",
"Eggs", "Honey Pots", "Total Drones")))
# Create the plot
ggplot(block_sum_long, aes(x = variable, y = mean_value, fill = block)) +
geom_col() +
theme_minimal() +
labs(title = "Mean Values of Brood Variables by Block",
x = "Variable",
y = "Mean Value") +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
scale_fill_viridis_d(option = "A") # Optional color scheme for differentiation
Round plot
adl_round_sum <- adl %>%
group_by(round) %>%
summarise(
mmax = mean(max_infc, na.rm = TRUE),
se_mmax = sd(max_infc, na.rm = TRUE) / sqrt(n()),
m_total_adl = mean(total.adl, na.rm = TRUE),
se_m_total_adl = sd(total.adl, na.rm = TRUE) / sqrt(n()),
m_first_adl = mean(days_to_adl, na.rm = TRUE),
se_m_first_adl = sd(days_to_adl, na.rm = TRUE) / sqrt(n()),
m_avg = mean(avg_infc, na.rm = TRUE),
se_m_avg = sd(avg_infc, na.rm = TRUE) / sqrt(n())
)
# Reshape data to include standard errors
adl_long <- adl_round_sum %>%
pivot_longer(
cols = starts_with("m"), # Select mean columns
names_to = "metric", # Create a column for metric names
values_to = "mean_value" # Create a column for mean values
) %>%
mutate(
se_value = case_when(
metric == "mmax" ~ se_mmax,
metric == "m_total_adl" ~ se_m_total_adl,
metric == "m_first_adl" ~ se_m_first_adl,
metric == "m_avg" ~ se_m_avg,
TRUE ~ NA_real_
)
)
# Filter for m_avg and mmax
adl_long_avg_max <- adl_long %>%
filter(metric %in% c("m_avg", "mmax"))
# Plot with error bars
plot1 <- ggplot(adl_long_avg_max, aes(x = round, y = mean_value, fill = metric)) +
geom_col(position = position_dodge(0.7), width = 0.7) +
geom_errorbar(
aes(ymin = mean_value - se_value, ymax = mean_value + se_value),
position = position_dodge(0.7), width = 0.2
) +
labs(
x = "Round",
y = "Cells/ul",
fill = NULL
) +
scale_fill_manual(
values = c("m_avg" = "purple", "mmax" = "skyblue"),
labels = c("m_avg" = "Average Infection Level", "mmax" = "Maximum Infection Level")
) +
theme_minimal() +
theme(
axis.title = element_text(size = 14),
axis.text = element_text(size = 12),
legend.text = element_text(size = 12),
legend.position = "top"
)
# Filter for m_total_adl and m_first_adl
adl_long_adl <- adl_long %>%
filter(metric %in% c("m_total_adl", "m_first_adl"))
# Plot with error bars
plot2 <- ggplot(adl_long_adl, aes(x = round, y = mean_value, fill = metric)) +
geom_col(position = position_dodge(0.7), width = 0.7) +
geom_errorbar(
aes(ymin = mean_value - se_value, ymax = mean_value + se_value),
position = position_dodge(0.7), width = 0.2
) +
labs(
x = "Round",
y = "Days",
fill = NULL
) +
scale_fill_manual(
values = c("m_total_adl" = "tomato", "m_first_adl" = "gold"),
labels = c(
"m_total_adl" = "Number of Days Infection was Detected",
"m_first_adl" = "Days Until First Infection Detected"
)
) +
theme_minimal() +
theme(
axis.title = element_text(size = 14),
axis.text = element_text(size = 12),
legend.text = element_text(size = 12),
legend.position = "top"
)plot_grid(plot1, plot2, ncol=2, nrow =1)
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