due 9 April 2020

Wrangling data files and exploring data to identify potential problems are critical and time-consuming initial steps in any data analysis. Data scientists spend considerable time on this effort. In this homework assignment, you will wrangle some messy data files and seek to identify potential problems.

In the zip file hw3_rawdata.zip there is a README.txt file plus a set of five Excel files with data for a set of intercross mice. The mice are the same as those in Broman et al. (2015) doi:10.1534/g3.115.019778; these are the clinical phenotypes.

1. Combine the phenotype data into a single dataset with mice as rows and measurements as columns. See the README.txt file for details on which columns to focus on.

   Note that the mouse identifiers appear in several forms. Also note that, to merge the data from final_rbm_data.xls, you need to use the sample/mouse key in RBM_Tube_Number_Key.xls.

2. Explore the data for possible problems: typos, errors, outliers? If a particular trait appears in multiple files, check that the values correspond.

3. Write a report describing:

   • your approach for merging the files, and the challenges involved
   • your data diagnostics methods
   • a summary of the problems you found. (Don’t show everything you did, but just some representative graphs.)