2.dplyr-intro-exercises.Rmd

---
title: "Title"
author: "Your Name"
date: '`r format(Sys.time(), "Last modified: %d %b %Y")`'
output: html_document
---

## Tidy data

Read the simulated clinical dataset and put into the tidy form. Also load the `tidyr` package that we are going to use.
- Can you make a boxplot to visualise the effect of the different treatments?

```{r}
library(tidyr)
messyData <- read.delim("clinicalData.txt")
```

```{r}
## Your answer here

```


## The patients dataset

First load the file using the `read.delim` function. Also load the `dplyr` library that we will need later

```{r}
library(dplyr)

patients <- read.delim("patient-data.txt")
patients <- tbl_df(patients)
```

## Exercise: select

- Print all the columns between `Height` and `Grade_Level`
- Print all the columns between `Height` and `Grade_Level`, but NOT `Pet`
- Print the columns `Height` and `Weight`
    + try to do this without specifying the full names of the columns
- (OPTIONAL)
- Print the columns in alphabetical order
- Print all the columns whose name is less than 4 characters in length

There are several ways to solve these. Feel free to explore

```{r}
### Your answer here

```


```{r}
library(stringr)
patients_clean <- mutate(patients, Sex = factor(str_trim(Sex)))
patients_clean <- mutate(patients_clean, Height= as.numeric(str_replace_all(Height,pattern = "cm","")))
```


- For a follow-on study, we are interested in overweight smokers
    + clean the `Smokes` column to contain just `TRUE` or `FALSE` values
- We need to calculate the Body Mass Index (BMI) for each of our patients
- $BMI = (Weight) / (Height^2)$ 
    + where Weight is measured in Kilograms, and Height in Metres
- A BMI of 25 is considered overweight, calculate a new variable to indicate which individuals are overweight
- (EXTRA) What other problems can you find in the data?


```{r}
### Your answer here

```