library(tidyverse)
ins <- read_csv("https://denvirlab.marshall.edu/BMR617-2022/data/InsulinSensitivityBorkman.csv")

ggplot(ins, aes(y=InsulinSensitivity, x=1)) + geom_boxplot() +
  geom_point(position=position_jitter(width=0.1)) +
  ylab("Insulin Sensitivity (mg/m2/min)") +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank())

ggplot(ins, aes(y=`%C20-22`, x=1)) + geom_boxplot() +
  geom_point(position=position_jitter(width=0.1)) +
  ylab("%C20-22 Polyunsaturated Fatty Acids") +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank())  

ggplot(ins, aes(x=`%C20-22`, y=InsulinSensitivity)) + geom_point() +
  xlab("%C20-22 Polyunsaturated Fatty Acids") +
  ylab("Insulin Sensitivity (mg/m2/min)")

r <- cor(pull(ins, `%C20-22`), pull(ins, InsulinSensitivity))
r
r**2

r52 <- -.29
r52**2

# A closer look at the Triglycerides-Glucose data
met <- read_csv("https://denvirlab.marshall.edu/BMR617-2022/data/TH-B6-metabolic.csv")
met <- separate(met, MouseID, sep="-", into=c("Strain", "Diet", "ID"))
r <- cor(pull(met, TG), pull(met, Glucose))
r

ggplot(met, aes(x=TG, y=Glucose)) + 
  geom_point(aes(color=Diet, shape=Strain))