Simontuk / pagoda2

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Overview of pagoda 2

Demo Web Application

10X sorted PBMCs https://tinyurl.com/pagoda2demo

10X PBMC Dataset http://pklab.med.harvard.edu/nikolas/pagoda2/frontend/current/pagodaURL/index.html?fileURL=http://pklab.med.harvard.edu/nikolas/pagoda2/staticDemo/10xPBMC.bin

Installation Instructions

Installation of pagoda2 requires installation of the Non-metric space library (nmslib) and a custom built largeVis package. Pagoda 2 has been tested with nmslib v1.5.3 although it should work with any subsequent version. To install pagoda2 please follow the instructions below.

Install system dependencies for building nmslib, example here provided for Ubuntu

$ sudo apt-get update
$ sudo apt-get install build-essential cmake gsl-bin libgsl0-dev libeigen3-dev libboost-all-dev

# Download and build nmslib v1.5.3
# You will need to keep this directory after the installation
# so choose a location that will not be altered
# something like ~/lib/nmslib/ will be suitable
$ cd /path/of/choice
$ wget https://github.com/searchivarius/nmslib/archive/v1.5.3.tar.gz
$ tar xvzf v1.5.3.tar.gz
$ cd nmslib-1.5.3
$ cd similarity_search
$ cmake .
$ make

Install custom largeVis package.

# You just need a temporary directory for this. ( e.g. ~/tmp/largeVisCustom/)
$ cd /other/path/of/choice
$ wget http://pklab.med.harvard.edu/pagoda2dependencies/largeVis.tar.gz
$ sha1sum largeVis.tar.gz # (optional) should be: 0eb51b7322d795f3f6cc35aec03e5bdd3189fa1e
$ tar xvzf largeVis.tar.gz
# Start R and install largeVis
$ R
> library(devtools)
> install('largeVis/')
> q()

Install pagoda2

# Specify the NMSLIB_PATH enviroment to the location that you installed
# and built nmslib above
$ export NMSLIB_PATH=/path/to/nmslib-1.5.3/
# Start R
$ R
# Install Bioconductor dependencies
> source("http://bioconductor.org/biocLite.R")
> biocLite(c("GO.db", "org.Hs.eg.db"), suppressUpdates=TRUE)
# Install pagoda
> library(devtools)
> install_github("hms-dbmi/pagoda2")
> library('pagoda2')
# Pagoda2 is now ready to use

Install padoga2 as a docker container

Instructions for installing docker container with pagoda2

Basic analysis walkthrough

PCA-based basic walkthrough

Basic analysis with generation of pagoda2 web application

Pagoda2 Web Application analysis

About


Languages

Language:C++ 71.0%Language:JavaScript 18.5%Language:R 7.9%Language:CMake 0.8%Language:C 0.8%Language:HTML 0.6%Language:Makefile 0.4%Language:CSS 0.0%Language:Shell 0.0%