# Load necessary libraries
library(dplyr)

# Set seed for reproducibility
set.seed(123)

# Number of subjects
n_subjects <- 30

# Function to generate synthetic data
simulate_data <- function(n_subjects) {
  # Data 1: Fiber type distributions
  fiber_data <- data.frame(
    Subject_ID = 1:n_subjects,
    Total_Fibers = sample(200:2000, n_subjects, replace = TRUE),
    Fiber_Type_1_Percentage = round(rbeta(n_subjects, 4, 3) * 100, 1) # Beta distribution for a heavier Type 1 split
  ) %>%
    mutate(
      Fiber_Type_2_Percentage = 100 - Fiber_Type_1_Percentage,
      Fiber_Type_1_Count = round(Total_Fibers * Fiber_Type_1_Percentage / 100),
      Fiber_Type_2_Count = Total_Fibers - Fiber_Type_1_Count
    )
  
  # Data 2: 3000m running time trials (pre, middle, post)
  trial_data <- data.frame(
    Subject_ID = rep(1:n_subjects, each = 3),
    Time_Point = rep(c("Pre", "Middle", "Post"), times = n_subjects)
  ) %>%
    group_by(Subject_ID) %>%
    mutate(
      Pre_Time = round(rnorm(1, mean = 650, sd = 50), 0),
      Running_Time = case_when(
        Time_Point == "Pre" ~ Pre_Time,
        Time_Point == "Middle" ~ max(round(rnorm(1, mean = Pre_Time * 0.9, sd = 30), 0), 500),
        Time_Point == "Post" ~ max(round(rnorm(1, mean = Pre_Time * 0.8, sd = 30), 0), 500)
      )
    ) %>%
    ungroup() %>%
    select(-Pre_Time)
  
  # Data 3: Delta efficiency
  delta_efficiency_data <- data.frame(
    Subject_ID = 1:n_subjects,
    Delta_Efficiency = round(runif(n_subjects, min = 15, max = 30), 1) # Random values in a reasonable range
  )
  
  # Merge all data into one data frame
  merged_data <- fiber_data %>%
    left_join(trial_data, by = "Subject_ID") %>%
    left_join(delta_efficiency_data, by = "Subject_ID")
  
  return(merged_data)
}

# Generate the simulated data
simulated_data <- simulate_data(n_subjects)