Hannah Kalvin's INFO 550 Project

For my project, I will analyze the a simulated dataset similar to the structure of SEER. The data contains age-adjusted breast cancer mortality rates by state and race.

Conduct analysis using Docker

You will first need to pull the docker container in order to run the analysis and produce the report.

First fork and clone the git repository to your a folder in the root directory on your local computer. The code below assumes that your local folder in the root directory is named INFO550Project - make sure to update to reflect the name of your folder accordingly.

Then, to pull the docker container, you can use the following code in your terminal:

docker pull hkalvin/infoproj

Next, create a new folder for the report output within the INFO550Project folder, which contains the information you cloned from the github repository. You may use the code below to create this folder.

cd INFO550Project/
mkdir output

Then run docker image to produce the report. The code below assumes that the folder where you cloned this directory is 1) located in the root folder and 2) is named INFO550Project. If you change either of these, please be sure to update accordingly the file path in the portion of the bash code below before the colon.

docker run -v ~/INFO550Project:/project hkalvin/infoproj

If you do not have Docker, the information below details how to create the report within terminal

To analyze the data you will need to install some R packages.

The required R packages can be installed using make install in terminal once you have navigated to the project folder.

make install

Execute the analysis

Packages must be installed prior to creating the report (i.e. run make install before make report). You will notice that the repo contains some empty folders. This is purposeful as you can run the commands below in order to create the files that belong in each of those empty folders.

To create the analysis report, from the project folder you can run

make report

This will create a file called HKalvin_report.html which will output in your directory that contains the results.

Additional helpful information

To clean the data, from the project folder you can run

make clean_data/all2.csv
make clean_data/ds2_fig.csv

All2.csv will be used to generate the report tables and ds2_fig.csv will be used to generate the figures in the report.

Both cleaned analysis datasets will be saved to the clean_data folder within the project folder.

To generate all four figures used in the analysis report, you can run

make figures/%.png

For more information on the Makefile targets, from the project folder you can run

make help