dipPeak

DipPeak is a kernel density estimator for high-throughput sequencing data. Please see the R documentation for additional details.

################################################################################
#HOW TO INSTALL
dipPeak relies on 2 other R packages I wrote. You can get the latest source from github (these are public). 

#There are 2 ways to install dipPeak and its dependencies.
#1. The easy way: install directly from github into R:
require(devtools)
install_github("sheffien/IRangeKernels") #public
install_github("sheffien/CWriteR") #public
install_github("sheffien/dipPeak", "YOUR_USERNAME", auth_token="YOUR PAT") #private
#Create your PAT (personal access token) here: https://github.com/settings/applications

#2. The hard way: you can clone the repos and then install locally, instead of directly from github. Here are some basic commands to go this route:

#Get public repos:
git clone http://github.com/sheffien/IRangeKernels.git
git clone http://github.com/sheffien/CWriteR.git

#then get the dipPeak source. (This is still private so you will have to have an ssh key set up to clone in this way).
git clone git@github.com:sheffien/dipPeak.git

#Once you have these, start R and install them.
R
install.packages("path/to/IRangeKernels", repos=NULL)
install.packages("path/to/CWriteR", repos=NULL)
install.packages("path/to/dipPeak", repos=NULL)

################################################################################
#HOW TO UPDATE
#If you installed by cloning repos, can update later with git pull:
cd dipPeak
git pull

#If you installed directly from github with install_github, just reinstall to update.

################################################################################
#HOW TO RUN
#You have 2 options: Use a helper script to run on the command line, or run from within R directly.
#1 ON THE COMMAND LINE 
#To use the command-line runner, you must have the R package "optparse" installed.
install.packages("optparse")

#You can then run it on the command line like this to see options:
./dipPeakRunner.sh

#2 DIRECTLY FROM WITHIN R
#You can load the package and run from within R like so:
library(dipPeak)
?dipPeaks

################################################################################
#Some Misc Documentation:
#By default, dipPeak estimates the peak detection cutoff
#using all data, instead of a different cutoff based on each
#individual chromosome. This is highly recommended, since sometimes single 
#chromosome cutoffs can lead to weird results.