2015_SISBID_5_02_homework.Rmd

---
title: "RR Short Course Homeworks"
author: "Keith Baggerly"
date: ''
output:
  html_document:
    number_sections: yes
  pdf_document: default
---

# Day 1

## Part 1

Write a minimal toy R package
  
- add a README
- add a vignette
- add some data
- Can you write a minimal R package in 15 minutes or less?
  
## Part 2

Obtain the mutation information for your
favorite gene from the [TumorPortal](http://www.tumorportal.org/)
  
- do the mutation patterns differ by known subtypes? 
  Some of these include estrogen receptor (ER) status in breast cancer (BRCA),
  HER2 status in BRCA, tissue source site in sarcoma (SARC), and 
  serous/endometriod status in uterine cancer (UCEC).
- Are there other features of the data you'd like clarification on?
- write up your analysis as a vignette
- include your analysis and the TumorPortal data in 
  an R package

## Part 3

(Evening) Read [Potti et al (2006)](http://www.nature.com/nm/journal/v12/n11/abs/nm1491.html), [our letter](http://www.nature.com/nm/journal/v13/n11/full/nm1107-1276b.html) and [the response](http://www.nature.com/nm/journal/v13/n11/full/nm1107-1277.html), and [Brad Perez's letter](http://www.cancerletter.com/articles/20150109_1).
  
- What would you do? Discuss. 


# Day 2

## Part 1

Put your toy package in a git repository

- put the repository on github  
- clone your neighbor's repository
- add a new function to your neighbor's package
- submit a pull request to add your package to 
  his github repository
- incorporate your neighbor's suggested functions
  into your package
  
## Part 2

- repeat part 1 with your mutation analysis
- do the mutation patterns differ by [run batch](https://tcga-data.nci.nih.gov/tcgafiles/ftp_auth/distro_ftpusers/anonymous/other/metadata/metadata.current.txt)?
- post it to github
- read and comment on your neighbor's analysis
- are there sanity checks you might employ?
  
## Part 3
  
(Evening) Read [Petricoin et al (2002)](http://www.sciencedirect.com/science/article/pii/S0140673602077462), our [Bioinformatics paper](http://bioinformatics.oxfordjournals.org/content/20/5/777.long), and [Leek et el (2010)](http://www.nature.com/nrg/journal/v11/n10/full/nrg2825.html).
  
- Where have you encountered batch effects?
- Are there ways in testing for batch effects?
- What sanity checks do you employ regularly?
  
# Day 3

Structure your Tumor Analysis _as a report_, and include your report in your package as a vignette
  
- Include an Executive Summary
- Include a Figure
- What do positive and negative numbers mean?
- include sessionInfo()
- post it to github
- read and comment on your neighbor's report