Live_session_Part1.Rmd

---
title: "Live_Part1"
author: "Austin Nuckols"
date: "11/16/2020"
output: html_document
---

## Intro

These are notes

Shortcuts: 
To insert a chunk - Ctrl + Alt + I
To run code - Ctrl + Enter

```{r}
y <- 54
x + y

```

Using comments

```{r}
# this is comment
x + 20 # This is important
```
Shortcut:
Assignment - Alt + - (minus sign)

```{r}
z <- 100
```

## Practical Data Examples

### Up top

```{r}
install.packages("tidyverse", "lubridate", "here", "xlsx")
```

Package: Group of functions or useful data that a person will "write" -- these make it easier code

```{r}
library(tidyverse)
library(here)
```

### Mice

Table of mice data
- Anxiety treatments: A, B, Placebo
- Anxiety levels: Open field test
-- if a mouse is less anxious, then they will spend more time in the center

```{r}
library(xlsx)
```

Read the data

```{r}
mice <- read.xlsx(file = "data/part1/mice.xlsx",
          sheetIndex = 1)
```

```{r}
mice <- read_csv(file = "data/part1/mice.csv")
```

View your data 

```{r}
head(mice, 10)
view(mice)
mice

############

str(mice)
#tibble == data.frame for our purposes

summary(mice)
```

Manipulating the data

Shortcut: 
Pipe - %>% 
Ctrl + Shift + M

```{r}
mice %>% # "And then"
  select(-X1) %>% # Now change sex and treatment to factors
  mutate(sex = as.factor(sex),
         treatment = as.factor(treatment))
```


Soem of the mice were injured. We don't want these mice -- need to filter out these

mouse id = 1, 24, 37

```{r}
library(lubridate)

injured <- c(1, 24, 37)

mice_clean <- mice %>% # %>% is the same as "And then"
  select(-X1) %>% # Now change sex and treatment to factors
  mutate(sex = as.factor(sex), #mutate() adds/changes variables in a data frame
         treatment = as.factor(treatment)) %>% 
  #filter(weight > 20)
  filter(!mouse_id %in% injured) %>% # filter() removes or keeps rows based on logical criteria
  mutate(birthdates = mdy(birthdates))
```

Summary of the data

```{r}
mice_clean %>% 
  summary()

mice_clean %>% 
  group_by(sex) %>%  #group_by() allows you to define grouping variables after which mutate()'s and summarise()'s will be done with respect to those variables
  summarise(ave_weight = mean(weight), #summarise reduces your data down to the summary functions (mean, sd, median, etc.) that you choose
            med_weight = median(weight),
            sd_weight = sd(weight))

## gropu by two variables


mice_clean %>% 
  group_by(sex, treatment) %>% 
  summarise(ave_weight = mean(weight),
            med_weight = median(weight),
            sd_weight = sd(weight))

head(mice_clean)
```

- Graph 

ggplot2 is a package in tidyverse

On ggplot:
  The "gg" stands for Grammar of Graphics (https://vita.had.co.nz/papers/layered-grammar.html)
  
  In short, the grammar of graphics states that graphs are generated layer by layer, with elements being stacked on top of one another.

```{r}
mice_weight_gg <- ggplot(data = mice_clean,
       aes(x = birthdates,#aes() - where you define your variabels of interest
           y = weight,
           color = sex)) + 
  geom_point(alpha = 0.5) +  #alpha is for transparency
  geom_smooth(method = "lm", #geom_smooth is for adding lines of best fit
              formula = "y ~ x",
              se = FALSE) + #for nonlinear lines, use "nls" -- look this up
  theme_bw()+ #theme_*()'s are used to change the layout of your graph
  labs(x = "Birthdates", #Change your labels easily with labs()
       y = "Weight (g)",
       color = "Sex",
       title = "Mouse weights by sex over time")

mice_weight_gg

#Save the graph 

ggsave(filename = "results/mice_weight_over_time.png") #make sure to add file extension

#you can change the dimensions inside of the ggsave function
```

```{r}
mice_clean %>% 
  ggplot(data = .,
         aes(x = treatment,
             y = open_field_time*60))+

#BOXPLOT - MEDIAN_IQRs_RANGES
  #geom_boxplot()+
  
#BAR PLOT w/ ERRORBARS - MEAN_SD
  # geom_bar(stat = "summary",
  #          fun.y = "mean")+
  # 
  # stat_summary(geom = "errorbar",
  #              fun.data = "mean_sdl",
  #              fun.args = list(mult = 1),
  #              width = 0.25)+

#CROSSBAR PLOT - MEAN_SD
  stat_summary(geom =
                 "crossbar"
                 ,
               fun.data = 'mean_sdl',
               fun.args = list(mult = 1),
               #width = 0.25
               )+
  
  geom_point(alpha = 0.3)+
  theme_light()+
  labs(x = "Treatment",
       y = "Seconds spent in the center",
       title = "Treatment A and B are higher than Placebo")
```