Story
Little bunny Foo Foo
Went hopping through the forest
Scooping up the field mice
And bopping them on the head
Session 4
August 8, 2018
Review
dplyr
In the last session we talked about 5 important dplyr verbs. What do each of these functions do?
Answer
- Pick out observations by their values
Answer
- Reorder the rows
Answer
- Pick out variables by their names
Answer
- Create new variables using existing variables
Answer
- Collapse many values into a single summary
What are the three “rules” of dplyr verbs?
Answer
dplyr functions are written as verbs with the following rules:
The first argument is a data frame.
Inside the function, you can use column names as if they were variables.
- The output is a new data frame.
The Pipe Operator
The %>% operator lets you chain together processing steps.
Iterative
Repetitive
Chained
Piped!
Try it
Rewrite the following using pipes.
ex1 <- group_by(example, site_code)
ex2 <- mutate(ex1, tumor_size = tumor_size * 10)
ex3 <- summarize(ex2, avg_tumor = mean(tumor_size))Want to learn more about the pipe operator? The Pipes in R Tutorial from DataCamp is a great place to start!
Or read the chapter on Pipes in R4DS. (The source of this example.)
Quick Review
Source the example_single_patient.R file we created in our last session so that the example data frame and dplyr are loaded.
example:
| patient_id | age_dx | age_visit | tumor_size | site_code |
|---|---|---|---|---|
| 5554321 | 54 | 54 | 9.5 | C220 |
| 5554321 | 54 | 55 | 9.5 | C400 |
| 5554321 | 54 | 56 | 9.7 | C412 |
| 5554321 | 54 | 57 | 9.9 | C220 |
| 5554321 | 54 | 58 | 10.1 | C400 |
Review Task
Use the pipe operator %>% and dplyr verbs to complete the following task:
Use the
exampledatasetFilter out tumors larger than 10
Calculate
follow_uptime as the number of years between diagnosis and the patient’s visitCalculate the average follow up time by
site_code
Overview
Import
Hands-on Working with Data
Export
Importing Data
Gather your “exported” data
Download the CancerRegistryDataPatients.xlsx and CancerRegistryDataPatients.csv data files from the Session 4 Materials page.
In the Files pane of your RStudio project, click ▸ to open the folder in a file browser. Copy or move the data files from your downloads folder into your project folder. Check that you see the files in RStudio.
If you’re running RStudio in your browser (e.g. on rstudio.cloud), you can run the download.file() commands in the materials page.
Import Using RStudio
There are two main packages for reading in data: readr for CSV and readxl for Excel.
If you have the option in your data source, choose to export your data as a CSV file. The options and control for importing into R using readr are more powerful and more easily customized.
Import Excel File
Import CSV File
A Note About Dates
Here is a small reference for date/time formats. See ?strftime for more time format placeholders.
%m/%d/%y: 01/02/06%m/%d/%Y: 01/02/2006%Y-%m-%d: 2006-01-02%m/%d/%y %H:%M: 01/02/06 15:04%m/%d/%y %H:%M:%S: 01/02/06 15:04:05%m/%d/%y %I:%M:%S %p: 01/02/06 03:04:05 PM
Explore the Data!
Use RStudio to look through the dataset using the RStudio viewer. Try the functions View() and glimpse().
Brainstorming…
What questions can you ask and answer using this dataset and dplyr?
Export The Data
Export the data using write_csv()
To export the data, use the function write_csv() from readr. This function takes the data frame as the first object, and a path (or filename) where the file should be written.
It’s a good idea to save your output files in a separate directory, like output_04 for outputs from this session. If you get an error like this
Warning in open.connection(path, "wb"): cannot open file 'output_04/
CancerRegistryPatients_export.csv': No such file or directoryError in open.connection(path, "wb"): cannot open the connectionit is most likely because the directory output_04 doesn’t exist in your current working directory. You can use the Files pane in RStudio to create one, or you can run
Exercises
The Session 04 materials page includes a file called cancer_registry_dplyr.R. Download this file and move it into your project folder.
Or download it in R
I’ve added code to get you started, but there are pieces missing. Fill in the places where you see ________.