TCGA_LUAD_demographic.Rmd

---
title: "TCGA_LUAD Demographic Analysis"
author: "Hsiao-Yu Peng"
date: "2023-12-14"
output: pdf_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)

library(readr)
library(dplyr)
library(ggplot2)
```


```{r}
# Load data
lung_phe <- read_tsv("~/Desktop/methylation/lung/TCGA-LUAD.GDC_phenotype.tsv")

dim(lung_phe)
head(lung_phe)
```


## Explore Data Analsyis

### Gender Distribution
```{r}
gender_dist <- lung_phe %>% 
  group_by(gender.demographic) %>% 
  summarize(number = n())

print(gender_dist)

ggplot(gender_dist, aes(x = gender.demographic, y = number, fill = gender.demographic)) +
  geom_bar(stat = "identity") +
  labs(title = "Gender Distribution",
       x = "Gender",
       y = "Number") +
  theme_classic()
  
```


TCGA_LUAD data set has more female individuals than male. Because we focus on studying female nonsmoker with LUAD, we select female for further investigation.


### Smoking History Distribution
```{r}
# Select female and gropy by smoking history
smoking_dist <- lung_phe %>% 
  filter(gender.demographic == "female") %>% 
  group_by(tobacco_smoking_history) %>% 
  summarize(number = n())

print(smoking_dist)

# Create bar plot 
ggplot(smoking_dist, aes(x = tobacco_smoking_history, y= number, fill = tobacco_smoking_history )) +
  geom_bar(stat = "identity") +
  labs(title = "Female Smoking Distribution",
       x = "Smoking History",
       y = "Number") + 
  theme_classic()

```


Category "1" of Smoking History means non-smokers, others are smokers. We continue to study race distribution of female nonsmokers and smokers


### Race Distribution
```{r}
# Select female nonsmokers
race_dist <- lung_phe %>% 
  filter(gender.demographic == "female",
         tobacco_smoking_history == 1) %>% 
  group_by(race.demographic) %>% 
  summarize(number = n())

print(race_dist)

ggplot(race_dist, aes(x = race.demographic, y = number, fill = race.demographic)) +
  geom_bar(stat = "identity") +
  labs(title = "Race Distribution of Nonsmokers") +
  theme_classic() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))


# Select female smokers
race_dist <- lung_phe %>% 
  filter(gender.demographic == "female",
         tobacco_smoking_history != 1) %>% 
  group_by(race.demographic) %>% 
  summarize(number = n())


print(race_dist)

ggplot(race_dist, aes(x = race.demographic, y = number, fill = race.demographic)) +
  geom_bar(stat = "identity") +
  labs(title = "Race Distribution of Smokers") +
  theme_classic() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))


```


We only study race regarding Asian, Black or African American and White. Next step we explore age distribution.


### Non-smoker Age Distribution
```{r}
# Select female nonsmokers
nonsmokers_age <- lung_phe %>% 
  filter(tobacco_smoking_history ==1,
         gender.demographic == "female",
         race.demographic %in% c("asian", "black or african american", "white", "not reported")) %>% 
  select(submitter_id.samples, age_at_index.demographic, gender.demographic, 
         race.demographic, tobacco_smoking_history) 

# Find non-smokers' age range
nonsmokers_age %>% 
  group_by(race.demographic) %>% 
  summarize(age_range = paste(min(age_at_index.demographic, na.rm = T),
                              max(age_at_index.demographic, na.rm = T),
                              sep = " - "),
            count = n())


ggplot(nonsmokers_age, aes(x = race.demographic, y = age_at_index.demographic)) +
  geom_boxplot() + 
  geom_jitter(position = position_jitter(width = 0.2, height = 0), alpha = 0.3) +
  labs(title = "Age Distribution by Race for Smokers")

```


### Smokers Age Distribution
```{r}
smokers_age <- lung_phe %>% 
  filter(tobacco_smoking_history !=1,
         gender.demographic == "female",
         race.demographic %in% c("asian", "black or african american", "white")) %>% 
  select(submitter_id.samples, age_at_index.demographic, gender.demographic, 
         race.demographic, tobacco_smoking_history) 

# Find smokers' age range 
smokers_age %>% 
  group_by(race.demographic) %>% 
  summarize(age_range = paste(min(age_at_index.demographic, na.rm = T),
                              max(age_at_index.demographic, na.rm = T),
                              sep = " - "),
            count = n())


ggplot(smokers_age, aes(x = race.demographic, y = age_at_index.demographic)) +
  geom_boxplot() + 
  geom_jitter(position = position_jitter(width = 0.2, height = 0), alpha = 0.3) +
  labs(title = "Age Distribution by Race for Smokers")
```