NBLAST on-the-fly

This project provides a simple interactive demonstration of the NBLAST neuron search algorithm using the RStudio shiny web application framework for R.

Live demo

A live demo of this application is running at http://jefferislab.org/si/nblast/on-the-fly/

Running locally

You can actually run the nblast server on your own machine with relatively little effort. Instructions follow or watch the video run-through.

Pre-requsisites

• R > 3.1.1
• RStudio

Installation

Assuming that you have opened NBLAST_on-the-fly.Rproj

source("install_packages.R")

Data

You will also need to download our processed/registered version of the flycircuit.tw dataset. The flycircuit package will enable you to do this. Note that the data will be installed in a default location within the home folder of the current user; therefore you must download the data while running as the user that will be used to run the shiny app.

Assuming that you have opened NBLAST_on-the-fly.Rproj

source("download_data.R")

will do the trick.

Running Shiny

library(shiny)
runGitHub("jefferislab/NBLAST_on-the-fly")

If you want to hack NBLAST_on-the-fly:

1. git clone the repository.
2. Use RStudio to edit the code (recommended)
3. Run the shiny app with runApp

setwd("/path/to/NBLAST_on-the-fly")
library(shiny)
runApp()

You will need to run the app in browser such as Chrome that supports WebGL rather than in the RStudio browswer if you want to see the 3D content.

Setting up a server

If you want to set up a server of your own, perhaps to provide access to a new dataset, besides R and and the packages already mentioned you will need to install shiny server. There is a free, open source shiny server edition. Detailed instructions for different linux platforms are here:

http://www.rstudio.com/products/shiny/download-server/