Mean ±SE in scatterplot of three groups - r

Question

edit: It's occurred to me that standard error cannot / will not be computed by R for the non-HG species (in particular, GB) where n is rather small (n = 4 for GB). Even if this is not the case for my failure to produce the plots, this low n is likely to invalidate standard error calculations. I'd appreciate others' advice.
I've seen a few similar answers, but nothing that closely approximates to what I want to do.
I've successfully used the code located here -
https://community.rstudio.com/t/using-stat-instead-of-dplyr-to-summarize-groups-in-a-ggplot/13916/2
- to produce a biplot showing the mean +/- s.e. whiskers for two groups of samples.
However, I've not been successful in re-working this code for use when there are three groups of samples. Can anyone tell me where I'm going wrong and how to rectify it please?
Thank you very much indeed in advance for advice; much appreciated :-)
Sample data and code (below):
Species d13C lin d34S lin
GB -20.1 8.09
HG -20.1 8.09
HG -19.51 9.46
HG -17.55 16.81
HG -23.72 8.03
HG -23.79 9.02
HG -18.09 8.64
HG -21.98 10.95
HG -18.6 8.5
HG -22.75 10.9
HG -21.7 9.08
HG -25 7.23
HG -17.61 16.56
HG -20.5 10.17
HG -18.14 15.4
GB -22.01 15.78
GB -19.62 12.62
LB -21.69 14.96
LB -25.56 8.4
LB -23.43 13.61
LB -22.92 12.68
LB -23.81 15.47
LB -25.42 8.63
HG -26.11 14.15
HG -20.61 9.27
HG -19.18 15.67
HG -19.76 14.49
HG -17.98 8.77
HG -22.71 9.26
HG -23.33 9.8
HG -22.7 10.84
HG -25.05 14.48
HG -24.63 9.98
HG -21.08 10.53
HG -24.93 8.42
HG -19.62 15.6
HG -23.82 7.7
HG -20.08 8.83
HG -24.36 11.01
HG -24.55 9.25
HG -21.89 16.59
HG -19 14.9
HG -18.96 17.69
HG -24.49 7.73
HG -19.15 8.82
HG -21.68 15.42
HG -23.6 8.03
HG -21.07 7.7
HG -19.56 14.34
HG -18.31 7.19
HG -23.39 14.24
HG -19.48 17.21
HG -18.25 8.71
HG -23.9 12.06
HG -23.19 8.59
HG -25.03 15.09
HG -21.01 11.74
HG -20.44 14.47
HG -24.4 10.5
HG -24.17 13.62
HG -22.41 15.63
HG -23.23 15.03
HG -25.22 13.36
HG -25.74 7.21
HG -18.01 12.47
HG -19.51 13.72
HG -25 14.56
HG -23.8 8.04
HG -21.99 12.24
HG -19.43 15.52
HG -24.93 9.45
HG -24.86 9.27
LB -25.27 12.94
LB -25.58 12.73
HG -26.02 10.28
HG -24.02 15.89
GB -24.85 7.14
LB -23.24 16.17
library(pillar)
library(rlang)
library (dplyr)
library(ggplot2)
library(ggstance)
fnc = function(data, group, x, y, z, adj=1) {
group=enquo(group)
x = enquo(x)
y = enquo(y)
z = enquo(z)
# Set size of whisker end caps
wv = data %>% pull(!!y) %>% range %>% diff/100*adj
wh = data %>% pull(!!x) %>% range %>% diff/100*adj
# If grouping variable is numeric, turn it into a factor
if(data %>% pull(!!group) %>% is.numeric) {
data = data %>%
mutate(!!quo_name(group) := factor(!!group))
}
# Generate column names for the x and y means that we'll calculate below
xmean = paste0(quo_text(x), "_mean")
ymean = paste0(quo_text(y), "_mean")
zmean = paste0(quo_text(z), "_mean")
left_join(data,
data %>%
group_by(!!group) %>%
summarise_at(vars(!!x, !!y, !!z) funs(mean=mean))
) %>%
ggplot(aes(colour=!!group)) +
geom_point(aes(x=!!x, y=!!y, z=!!z), size=5) +
theme_classic() +
geom_point(aes(x=!!sym(xmean), y=!!sym(ymean), z=!!sym(zmean)), shape=18, size=3) +
stat_summary(fun.data=mean_se, fun.args=list(mult=1.96),
aes(x=!!sym(xmean), y=!!y, z=!!z),
geom="errorbar", width=wh) +
stat_summaryh(fun.data=mean_se_h, fun.args=list(mult=1.96),
aes(x=!!x, y=!!y, z=!!z, sym(ymean)),
geom="errorbarh", width=wv)
}
##plot the ERROR BAR
fnc(data, Species, d13C.lin, d34S.lin) +
theme(axis.text=element_text(size=12, face = "bold", colour = "black"),
axis.title=element_text(size=17,face="bold", colour = "black")) +
theme(legend.text=element_text(size=14, face = "bold", colour = "black")) +
theme(legend.background = element_rect(fill="white",
size=0.5, linetype="solid",
colour ="black")) +
theme(legend.title = element_text(size = 14, face = "bold", colour = "black")) +
labs(y = expression(delta^{34}*"S"*" (‰)"), x = expression(delta^{13}*"C"*" (‰)"))
This is the error message I receive:
Error in is_call(expr, paren_sym) :
argument "expr" is missing, with no default
As I say, any help greatly appreciated please.

Answer 1

The error comes from the z argument which is not passed as argument in the function.
fnc <- function(data, group, x, y, adj=1) {
group <- enquo(group)
x <- enquo(x)
y <- enquo(y)
# Set size of whisker end caps
wv <- data %>%
pull(!!y) %>%
range %>%
diff/100*adj
wh <- data %>%
pull(!!x) %>%
range %>%
diff/100*adj
# If grouping variable is numeric, turn it into a factor
if(data %>%
pull(!!group) %>%
is.numeric) {
data <- data %>%
mutate(!!quo_name(group) := factor(!!group))
}
# Generate column names for the x and y means that we'll calculate below
xmean <- str_c(as_name(x), "_mean")
ymean <- str_c(as_name(y), "_mean")
left_join(data,
data %>%
group_by(!!group) %>%
summarise_at(vars(!!x, !!y), list(mean = ~ mean(.)))) %>%
ggplot(aes(colour=!!group)) +
geom_point(aes(x=!!x, y=!!y), size=5) +
theme_classic() +
geom_point(aes(x=!!sym(xmean), y=!!sym(ymean)), shape=18, size=3) +
stat_summary(fun.data=mean_se, fun.args=list(mult=1.96),
aes(x=!!sym(xmean), y=!!y),
geom="errorbar", width=wh) +
stat_summaryh(fun.data=mean_se_h, fun.args=list(mult=1.96),
aes(x=!!x, y = !!y),
geom="errorbarh", width=wv)
}
-testing
fnc(df1, Species, d13C.lin, d34S.lin)
data
df1 <- structure(list(Species = c("GB", "HG", "HG", "HG", "HG", "HG",
"HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "GB", "GB",
"LB", "LB", "LB", "LB", "LB", "LB", "HG", "HG", "HG", "HG", "HG",
"HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG",
"HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG",
"HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG",
"HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG", "HG",
"LB", "LB", "HG", "HG", "GB", "LB"), d13C.lin = c(-20.1, -20.1,
-19.51, -17.55, -23.72, -23.79, -18.09, -21.98, -18.6, -22.75,
-21.7, -25, -17.61, -20.5, -18.14, -22.01, -19.62, -21.69, -25.56,
-23.43, -22.92, -23.81, -25.42, -26.11, -20.61, -19.18, -19.76,
-17.98, -22.71, -23.33, -22.7, -25.05, -24.63, -21.08, -24.93,
-19.62, -23.82, -20.08, -24.36, -24.55, -21.89, -19, -18.96,
-24.49, -19.15, -21.68, -23.6, -21.07, -19.56, -18.31, -23.39,
-19.48, -18.25, -23.9, -23.19, -25.03, -21.01, -20.44, -24.4,
-24.17, -22.41, -23.23, -25.22, -25.74, -18.01, -19.51, -25,
-23.8, -21.99, -19.43, -24.93, -24.86, -25.27, -25.58, -26.02,
-24.02, -24.85, -23.24), d34S.lin = c(8.09, 8.09, 9.46, 16.81,
8.03, 9.02, 8.64, 10.95, 8.5, 10.9, 9.08, 7.23, 16.56, 10.17,
15.4, 15.78, 12.62, 14.96, 8.4, 13.61, 12.68, 15.47, 8.63, 14.15,
9.27, 15.67, 14.49, 8.77, 9.26, 9.8, 10.84, 14.48, 9.98, 10.53,
8.42, 15.6, 7.7, 8.83, 11.01, 9.25, 16.59, 14.9, 17.69, 7.73,
8.82, 15.42, 8.03, 7.7, 14.34, 7.19, 14.24, 17.21, 8.71, 12.06,
8.59, 15.09, 11.74, 14.47, 10.5, 13.62, 15.63, 15.03, 13.36,
7.21, 12.47, 13.72, 14.56, 8.04, 12.24, 15.52, 9.45, 9.27, 12.94,
12.73, 10.28, 15.89, 7.14, 16.17)), class = "data.frame",
row.names = c(NA,
-78L))