Rstudio Research Agent

# R specific
.Rhistory
.RData
.Rproj.user

# renv
renv/library/
renv/staging/

# User-specific settings
.claude/settings.local.json

# ============================================================================
# Analysis Script Template
# ============================================================================
# Description: [Brief description of what this analysis does]
# Author: [Your name]
# Date: [Date]
# Input: data/raw/[input_file]
# Output: results/[output_files]
# ============================================================================

# Load libraries --------------------------------------------------------------
library(tidyverse)
library(here)

# Set options -----------------------------------------------------------------
options(scipen = 999)  # Disable scientific notation

# Define paths ----------------------------------------------------------------
DATA_RAW <- here("data/raw")
DATA_PROCESSED <- here("data/processed")
RESULTS_FIGURES <- here("results/figures")
RESULTS_TABLES <- here("results/tables")
RESULTS_MODELS <- here("results/models")

# Load data -------------------------------------------------------------------
# df <- read_csv(file.path(DATA_RAW, "filename.csv"))

# Data preparation ------------------------------------------------------------
# Clean and transform data here

# Analysis --------------------------------------------------------------------
# Perform your analysis here

# Visualization ---------------------------------------------------------------
# Create plots with ggplot2

# Save results ----------------------------------------------------------------
# ggsave(file.path(RESULTS_FIGURES, "figure1.png"), ...)

# Session info (for reproducibility) -----------------------------------------
sessionInfo()

# ============================================================================
# Publication-Quality Plot Examples
# ============================================================================
# Description: Example scripts demonstrating common publication plots
# Author: [Your name]
# Date: [Date]
# ============================================================================

library(ggplot2)
library(dplyr)
library(patchwork)

# Source the plot functions
source("templates/plot_functions.R")

# ============================================================================
# Example 1: Scatter Plot with Regression Line
# ============================================================================

# Simulated data
set.seed(42)
scatter_data <- data.frame(
  x = rnorm(100, mean = 50, sd = 10),
  y = rnorm(100, mean = 50, sd = 10)
)
scatter_data$y <- scatter_data$x * 0.8 + rnorm(100, sd = 5)
scatter_data$group <- sample(c("Control", "Treatment"), 100, replace = TRUE)

# Create scatter plot
p_scatter <- ggplot(scatter_data, aes(x = x, y = y, color = group)) +
  geom_point(size = 3, alpha = 0.7) +
  geom_smooth(method = "lm", se = TRUE, linewidth = 0.8) +
  scale_color_manual(values = c("#E69F00", "#0072B2")) +
  theme_publication() +
  labs(x = "Independent Variable (units)",
       y = "Dependent Variable (units)",
       color = "Group")

# Save
save_publication_figure(p_scatter, "scatter_regression",
                        width = 89, height = 89,
                        formats = c("pdf", "png"))

# ============================================================================
# Example 2: Bar Plot with Error Bars
# ============================================================================

# Summary data for bar plot
bar_data <- data.frame(
  category = c("A", "B", "C", "D"),
  mean_value = c(25, 40, 35, 50),
  se = c(3, 4, 3.5, 5)
)

# Create bar plot
p_bar <- ggplot(bar_data, aes(x = category, y = mean_value, fill = category)) +
  geom_bar(stat = "identity", width = 0.7) +
  geom_errorbar(aes(ymin = mean_value - se, ymax = mean_value + se),
                width = 0.2, linewidth = 0.8) +
  scale_fill_manual(values = okabe_ito[1:4]) +
  theme_publication() +
  labs(x = "Category",
       y = "Mean Value ± SE",
       fill = "Category") +
  theme(legend.position = "none")

# Save
save_publication_figure(p_bar, "bar_with_error",
                        width = 89, height = 89,
                        formats = c("pdf", "png"))

# ============================================================================
# Example 3: Volcano Plot (Differential Expression)
# ============================================================================

# Simulated differential expression data
set.seed(42)
n_genes <- 1000
volcano_data <- data.frame(
  gene = paste0("Gene", 1:n_genes),
  log2FC = rnorm(n_genes, 0, 2),
  pvalue = runif(n_genes, 0.0001, 1)
)

# Add significance
volcano_data <- volcano_data %>%
  mutate(
    neglog10p = -log10(pvalue),
    significance = case_when(
      log2FC >= 1 & pvalue < 0.05 ~ "Up",
      log2FC <= -1 & pvalue < 0.05 ~ "Down",
      TRUE ~ "NS"
    )
  )

# Create volcano plot
p_volcano <- ggplot(volcano_data, aes(x = log2FC, y = neglog10p)) +
  geom_point(aes(color = significance), alpha = 0.6, size = 1.5) +
  scale_color_manual(
    values = c("Down" = "#0072B2", "NS" = "#999999", "Up" = "#D55E00"),
    name = ""
  ) +
  geom_vline(xintercept = c(-1, 1), linetype = "dashed", color = "grey50") +
  geom_hline(yintercept = -log10(0.05), linetype = "dashed", color = "grey50") +
  theme_publication() +
  labs(x = expression(log[2]~Fold~Change),
       y = expression(-log[10]~P~Value)) +
  theme(legend.position = "right")

# Save
save_publication_figure(p_volcano, "volcano_plot",
                        width = 100, height = 89,
                        formats = c("pdf", "png"))

# ============================================================================
# Example 4: Box Plot
# ============================================================================

# Simulated data for box plot
set.seed(42)
box_data <- data.frame(
  group = rep(c("Control", "Treatment A", "Treatment B"), each = 30),
  value = c(rnorm(30, 20, 5), rnorm(30, 35, 8), rnorm(30, 30, 6))
)

# Create box plot
p_box <- ggplot(box_data, aes(x = group, y = value, fill = group)) +
  geom_boxplot(outlier.shape = NA, width = 0.7) +
  geom_point(position = position_jitter(width = 0.1),
             size = 1.5, alpha = 0.5) +
  scale_fill_manual(values = okabe_ito[1:3]) +
  theme_publication() +
  labs(x = "Group",
       y = "Measured Value",
       fill = "Group") +
  theme(legend.position = "none")

# Save
save_publication_figure(p_box, "box_plot",
                        width = 89, height = 89,
                        formats = c("pdf", "png"))

# ============================================================================
# Example 5: Multi-Panel Figure (2x2 grid)
# ============================================================================

# Create four different plots for demonstration
p1 <- ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point(color = okabe_ito[1], size = 2, alpha = 0.7) +
  geom_smooth(method = "lm", se = FALSE, color = "black") +
  theme_publication(base_size = 10) +
  labs(x = "Weight", y = "MPG", title = "A")

p2 <- ggplot(mtcars, aes(x = factor(cyl), y = hp, fill = factor(cyl))) +
  geom_boxplot(outlier.shape = NA) +
  geom_point(position = position_jitter(width = 0.1),
             size = 1, alpha = 0.5) +
  scale_fill_okabe_ito() +
  theme_publication(base_size = 10) +
  labs(x = "Cylinders", y = "Horsepower", title = "B") +
  theme(legend.position = "none")

p3 <- ggplot(diamonds[sample(nrow(diamonds), 1000), ],
             aes(x = carat, y = price)) +
  geom_point(color = okabe_ito[2], size = 1, alpha = 0.5) +
  theme_publication(base_size = 10) +
  labs(x = "Carat", y = "Price", title = "C")

p4 <- ggplot(mtcars, aes(x = gear, fill = factor(am))) +
  geom_bar(position = "fill") +
  scale_fill_okabe_ito() +
  theme_publication(base_size = 10) +
  labs(x = "Gear", y = "Proportion", fill = "Transmission", title = "D")

# Combine into 2x2 grid
combined_figure <- (p1 | p2) / (p3 | p4)

# Save multi-panel figure
ggsave("results/figures/multipanel_2x2.pdf",
       combined_figure,
       width = 183, height = 183,
       units = "mm", dpi = 600)

# ============================================================================
# Example 6: Line Plot (Time Series)
# ============================================================================

# Simulated time series data
set.seed(42)
time_points <- rep(1:10, 3)
line_data <- data.frame(
  time = time_points,
  value = c(
    20 + time_points[1:10] + rnorm(10, 0, 2),
    25 + time_points[1:10] * 1.2 + rnorm(10, 0, 3),
    15 + time_points[1:10] * 0.8 + rnorm(10, 0, 2)
  ),
  group = rep(c("Series A", "Series B", "Series C"), each = 10)
)

# Create line plot
p_line <- ggplot(line_data, aes(x = time, y = value, color = group)) +
  geom_line(linewidth = 0.8) +
  geom_point(size = 2.5) +
  scale_color_manual(values = okabe_ito[1:3]) +
  theme_publication() +
  labs(x = "Time (hours)",
       y = "Measured Value (arbitrary units)",
       color = "Group") +
  theme(legend.position = "top",
        panel.grid.minor = element_line(color = "grey95"))

# Save
save_publication_figure(p_line, "line_plot",
                        width = 120, height = 89,
                        formats = c("pdf", "png"))

# ============================================================================
# Example 7: Heatmap (Correlation Matrix)
# ============================================================================

# Create correlation matrix
corr_matrix <- cor(mtcars[, c("mpg", "cyl", "disp", "hp", "wt", "qsec")])

# Convert to long format
corr_long <- expand.grid(dimnames(corr_matrix)) %>%
  mutate(correlation = as.vector(corr_matrix),
         Row = Var1, Col = Var2)

# Create heatmap
p_heat <- ggplot(corr_long, aes(x = Col, y = Row, fill = correlation)) +
  geom_tile(color = "white", linewidth = 0.5) +
  scale_fill_gradient2(low = "#0072B2", mid = "white", high = "#D55E00",
                       midpoint = 0, limits = c(-1, 1)) +
  theme_publication() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(x = "", y = "", fill = "Correlation")

# Save
save_publication_figure(p_heat, "heatmap",
                        width = 120, height = 100,
                        formats = c("pdf", "png"))

# ============================================================================
# Session Info
# ============================================================================

sessionInfo()

# ============================================================================
# Publication-Quality Plot Functions
# ============================================================================
# Description: Collection of functions for creating publication-ready plots
# Author: [Your name]
# Date: [Date]
# ============================================================================

# Required libraries ----------------------------------------------------------
library(ggplot2)
library(scales)
library(patchwork)

# Color-blind friendly palettes ---------------------------------------------
# Okabe-Ito palette (recommended for publications)
okabe_ito <- c(
  "#E69F00", # Orange
  "#56B4E9", # Sky blue
  "#009E73", # Bluish green
  "#F0E442", # Yellow
  "#0072B2", # Blue
  "#D55E00", # Vermillion
  "#CC79A7", # Reddish purple
  "#999999"  # Grey
)

# Diverging palette for heatmaps/volcano plots
diverging_palette <- c("#D55E00", "#F0E442", "#999999", "#56B4E9", "#0072B2")

# ============================================================================
# Publication Theme
# ============================================================================

#' Publication-Ready ggplot2 Theme
#'
#' Creates a clean, publication-quality theme for ggplot2 figures
#' following common journal guidelines.
#'
#' @param base_size Base font size in points
#' @param base_family Base font family (Arial recommended for publications)
#' @return A ggplot theme object
#'
#' @examples
#' ggplot(mtcars, aes(x = wt, y = mpg)) +
#'   geom_point() +
#'   theme_publication()
theme_publication <- function(base_size = 12, base_family = "Arial") {
  theme_bw(base_size = base_size, base_family = base_family) +
  theme(
    # Legend placement
    legend.position = "top",
    legend.title = element_text(size = rel(1)),
    legend.text = element_text(size = rel(0.9)),
    legend.key.size = unit(0.4, "cm"),
    legend.margin = margin(0, 0, 0, 0, "cm"),
    legend.box.margin = margin(0, 0, -5, 0, "mm"),

    # Panel borders
    panel.border = element_rect(size = 0.5, color = "black"),
    panel.grid.major = element_line(color = "grey90", linewidth = 0.25),
    panel.grid.minor = element_blank(),

    # Axis text
    axis.text = element_text(size = rel(0.9), color = "black"),
    axis.title = element_text(size = rel(1)),

    # Plot elements
    plot.title = element_text(size = rel(1.2), hjust = 0.5),

    # Strip text for facets
    strip.background = element_rect(color = "black", fill = "grey90"),
    strip.text = element_text(size = rel(0.9))
  )
}

# ============================================================================
# Scale Functions
# ============================================================================

#' Okabe-Ito Color Scale
#'
#' Applies the color-blind friendly Okabe-Ito palette
#'
#' @param palette Number of colors to use (1-8)
#'
scale_color_okabe_ito <- function(palette = 8) {
  discrete_scale("colour", "okabe_ito",
                 function(n) okabe_ito[1:min(n, 8)])
}

scale_fill_okabe_ito <- function(palette = 8) {
  discrete_scale("fill", "okabe_ito",
                 function(n) okabe_ito[1:min(n, 8)])
}

# ============================================================================
# Export Functions
# ============================================================================

#' Save Publication-Quality Figure
#'
#' Exports a ggplot object to publication-ready formats with proper
#' dimensions and resolution.
#'
#' @param plot ggplot object to save
#' @param filename Output filename (without extension)
#' @param path Output directory path
#' @param width Width in mm
#' @param height Height in mm
#' @param dpi Resolution for raster formats (default: 600)
#' @param formats Vector of formats: c("pdf", "png", "svg", "tiff")
#'
#' @examples
#' p <- ggplot(mtcars, aes(x = wt, y = mpg)) + geom_point()
#' save_publication_figure(p, "scatter_plot",
#'                         width = 89, height = 89,
#'                         formats = c("pdf", "png"))
save_publication_figure <- function(plot,
                                    filename,
                                    path = "results/figures",
                                    width = 89,
                                    height = 89,
                                    dpi = 600,
                                    formats = c("pdf")) {

  # Create output directory if needed
  if (!dir.exists(path)) {
    dir.create(path, recursive = TRUE)
  }

  # Export in each requested format
  for (fmt in formats) {
    filepath <- file.path(path, paste0(filename, ".", fmt))

    ggsave(filepath, plot = plot,
           width = width, height = height,
           units = "mm", dpi = dpi,
           device = fmt)

    message("Saved: ", filepath)
  }
}

# ============================================================================
# Specific Plot Functions
# ============================================================================

#' Create Scatter Plot with Regression
#'
#' @param data Data frame
#' @param x x-axis variable
#' @param y y-axis variable
#' @param color Color/group variable (optional)
#' @param add_regression Add regression line
#' @param show_ci Show confidence interval
#'
create_scatter_plot <- function(data, x, y, color = NULL,
                                add_regression = TRUE, show_ci = TRUE) {
  x_sym <- enquo(x)
  y_sym <- enquo(y)

  p <- ggplot(data, aes(x = !!x_sym, y = !!y_sym))

  if (is.null(color)) {
    p <- p + geom_point(size = 3, alpha = 0.7)
    if (add_regression) {
      if (show_ci) {
        p <- p + geom_smooth(method = "lm", se = TRUE, color = "black")
      } else {
        p <- p + geom_smooth(method = "lm", se = FALSE, color = "black")
      }
    }
  } else {
    color_sym <- enquo(color)
    p <- p + geom_point(aes_string(color = color), size = 3, alpha = 0.7) +
      scale_color_okabe_ito()
    if (add_regression) {
      if (show_ci) {
        p <- p + geom_smooth(aes_string(color = color),
                            method = "lm", se = TRUE)
      } else {
        p <- p + geom_smooth(aes_string(color = color),
                            method = "lm", se = FALSE)
      }
    }
  }

  p + theme_publication()
}

#' Create Bar Plot with Error Bars
#'
#' @param data Data frame
#' @param x x-axis variable (categorical)
#' @param y y-axis variable (numeric)
#' @param fill Fill variable (optional)
#' @param error_var Error bar variable (optional)
#'
create_bar_plot <- function(data, x, y, fill = NULL, error_var = NULL) {
  x_sym <- enquo(x)
  y_sym <- enquo(y)

  p <- ggplot(data, aes(x = !!x_sym, y = !!y_sym, group = !!x_sym))

  if (is.null(fill)) {
    p <- p + geom_bar(stat = "identity", fill = okabe_ito[1], width = 0.7)
  } else {
    fill_sym <- enquo(fill)
    p <- p + geom_bar(aes_string(fill = fill),
                      stat = "identity", width = 0.7, position = "dodge") +
      scale_fill_okabe_ito()
  }

  if (!is.null(error_var)) {
    p <- p + geom_errorbar(aes_string(ymin = paste0(y, "-", error_var),
                                      ymax = paste0(y, "+", error_var)),
                          width = 0.2)
  }

  p + theme_publication() +
    labs(y = deparse(substitute(y)), x = deparse(substitute(x)))
}

#' Create Volcano Plot
#'
#' @param data Data frame with log2FC and p-value columns
#' @param logfc Column name for log2 fold change
#' @param pval Column name for p-value
#' @param fc_threshold Fold change threshold
#' @param p_threshold P-value threshold
#'
create_volcano_plot <- function(data, logfc = "log2FC", pval = "pvalue",
                                 fc_threshold = 1, p_threshold = 0.05) {
  # Calculate -log10(p-value)
  data <- data %>%
    mutate(neglog10p = -log10(!!sym(pval)),
           significance = case_when(
             abs(!!sym(logfc)) >= fc_threshold & !!sym(pval) < p_threshold ~ "Up",
             abs(!!sym(logfc)) >= fc_threshold & !!sym(pval) < p_threshold ~ "Down",
             TRUE ~ "NS"
           ))

  p <- ggplot(data, aes(x = !!sym(logfc), y = neglog10p)) +
    geom_point(aes(color = significance), alpha = 0.6, size = 1.5) +
    scale_color_manual(values = c("Down" = "#0072B2",
                                   "NS" = "#999999",
                                   "Up" = "#D55E00"),
                       name = "") +
    geom_vline(xintercept = c(-fc_threshold, fc_threshold),
               linetype = "dashed", color = "grey50") +
    geom_hline(yintercept = -log10(p_threshold),
               linetype = "dashed", color = "grey50") +
    theme_publication() +
    labs(x = expression(log[2]~Fold~Change),
         y = expression(-log[10]~P~Value))

  return(p)
}

#' Create Box Plot
#'
#' @param data Data frame
#' @param x x-axis variable (categorical)
#' @param y y-axis variable (numeric)
#' @param fill Fill variable (optional)
#'
create_box_plot <- function(data, x, y, fill = NULL) {
  x_sym <- enquo(x)
  y_sym <- enquo(y)

  p <- ggplot(data, aes(x = !!x_sym, y = !!y_sym, group = !!x_sym))

  if (is.null(fill)) {
    p <- p + geom_boxplot(fill = okabe_ito[1], outlier.shape = NA) +
      geom_point(position = position_jitter(width = 0.1), size = 1, alpha = 0.5)
  } else {
    fill_sym <- enquo(fill)
    p <- p + geom_boxplot(aes_string(fill = fill),
                          outlier.shape = NA) +
      geom_point(position = position_jitterdodge(dodge.width = 0.75),
                 size = 1, alpha = 0.5) +
      scale_fill_okabe_ito()
  }

  p + theme_publication()
}

# ============================================================================
# Session Info
# ============================================================================

# For reproducibility
sessioninfo::session_info()

# ============================================================================
# R Project Initialization Script
# ============================================================================
# This script sets up a new R research project with proper directory structure
# and package management using renv.
#
# Usage: Run this script at the start of a new project
# ============================================================================

# Setup message
message("Setting up R research project...")

# Create directory structure
dirs <- c(
  "data/raw",
  "data/processed",
  "scripts",
  "results/figures",
  "results/tables",
  "results/models",
  "reports",
  "docs"
)

for (dir in dirs) {
  if (!dir.exists(dir)) {
    dir.create(dir, recursive = TRUE)
    message("Created directory: ", dir)
  }
}

# Create .gitkeep files for empty directories
gitkeep_dirs <- c(
  "data/raw/.gitkeep",
  "data/processed/.gitkeep",
  "results/figures/.gitkeep",
  "results/tables/.gitkeep",
  "results/models/.gitkeep",
  "reports/.gitkeep"
)

for (file in gitkeep_dirs) {
  if (!file.exists(file)) {
    write("", file = file)
  }
}

# Initialize renv if not already initialized
if (!file.exists("renv.lock")) {
  if (interactive()) {
    response <- readline("Initialize renv for package management? (y/n): ")
    if (tolower(response) == "y") {
      renv::init()
      message("renv initialized. Remember to snapshot your packages:")
      message("  renv::snapshot()")
    }
  } else {
    renv::init()
    message("renv initialized. Remember to snapshot your packages:")
    message("  renv::snapshot()")
  }
}

# Create .Rproj file if using RStudio
if (!file.exists(paste0(getwd(), ".Rproj"))) {
  if (requireNamespace("rstudioapi", quietly = TRUE) && rstudioapi::isAvailable()) {
    rstudioapi::createProject(getwd())
    message("Created RStudio project file")
  }
}

# Create a basic README
if (!file.exists("README.md")) {
  readme_content <- c(
    "# Project Title",
    "",
    "## Description",
    "",
    "Brief description of the research project.",
    "",
    "## Project Structure",
    "",
    "```",
    ".",
    "|-- data/",
    "|   |-- raw/           # Original, immutable data",
    "|   |-- processed/     # Cleaned, transformed data",
    "|-- scripts/           # Analysis scripts",
    "|-- results/",
    "|   |-- figures/       # Plots and visualizations",
    "|   |-- tables/        # Summary tables",
    "|   |-- models/        # Saved model objects",
    "|-- reports/           # R Markdown/Quarto reports",
    "|-- docs/              # Additional documentation",
    "```",
    "",
    "## Setup",
    "",
    "```r",
    "# Install and activate renv",
    "renv::restore()",
    "```",
    "",
    "## Usage",
    "",
    "## References",
    ""
  )
  writeLines(readme_content, "README.md")
  message("Created README.md")
}

message("\n=== Project setup complete! ===")
message("\nNext steps:")
message("1. Add your data to data/raw/")
message("2. Create analysis scripts in scripts/")
message("3. Use renv::install() to add required packages")
message("4. Run renv::snapshot() to save your package state")

{
  "R": {
    "Version": "4.3.2",
    "Platforms": [
      {
        "Platform": "x86_64-pc-linux-gnu",
        "Path": "library"
      }
    ]
  },
  "Packages": {
    " tidyverse ": " {
      "Package": "tidyverse",
      "Version": "2.0.0",
      "Source": "Repository",
      "Repository": "CRAN",
      "Hash": "abc123..."
    },
    "ggplot2": {
      "Package": "ggplot2",
      "Version": "3.4.4",
      "Source": "Repository",
      "Repository": "CRAN",
      "Hash": "def456..."
    },
    "dplyr": {
      "Package": "dplyr",
      "Version": "1.1.3",
      "Source": "Repository",
      "Repository": "CRAN",
      "Hash": "ghi789..."
    }
  }
}

---
title: "Report Title"
author: "Author Name"
date: "`r Sys.Date()`"
format:
  html:
    theme: cosmo
    toc: true
    code-fold: true
    code-tools: true
  pdf:
    toc: true
    number-sections: true
engine: knitr
---

# Introduction

Brief description of the research question and objectives.

# Methods

## Data Preparation

```{r}
# Load libraries
library(tidyverse)
library(here)
```

Describe the data processing steps.

## Analysis

```{r}
# Your analysis code here
```

# Results

```{r}
# Results
```

# Discussion

Interpretation of results and conclusions.

# Session Info

```{r}
sessionInfo()
```

# References

---

## Appendices

### Additional Analyses

```{r}
# Additional code
```