Awesome

Intro

This is a little web app aiming to show the power of DNAism and bedserver for exploring genomic datasets.

Details

DNAism is a javascript library to visualize genomic data using Horizon Charts. Head to the DNAism page to get more details. The user defines the region of the genome to visualize by defining a context. Then it specifies sources and samples which are the components that provide the actual data. The component that does the plotting (horizon) uses the source and the samples components to pull the necessary datapoints for the visualization. In the context, we also define the size of the plot, the sizes determines how many datapoints the horizon component requests to the sources.

When I started writing DNAism, I created the most simple source posible: a bedfile. Although simple, it was inefficient since we had to pull all the available datapoints for a region in order to compute the values required by the visualization, typically much less than the size of the region.

That's why I wrote another source, bedserver, which let's a backend to do the heavy lifting and return to the client (browser) only the necessary points for the visualization. Besides the new source, I also wrote the (Bedserver), which is the backend for the bedserver source.

Example

Let me walk you through a simple example to see how all this projects work together to visualize bed datasets.

First, let's install the (Bedserver), follow the (README) to perform the installation and familiarize yourself with how the server works.

Now, let's download the webbroser webapp in the current directory. Along with the app, we are going to download a couple of testing datasets to be able to use the app. You can later add your own datasets:

$ cd /somewhere
$ curl -L https://github.com/drio/bedbrowser/tarball/master | tar zxvf -
$  cd drio-bedbrowser-*

Now fire up the bedserver:

$ bedserver
 * Running on http://127.0.0.1:5000/
 * Restarting with reloader

And, now fire up your browser and go to localhost:5000/index.html. You should see something like this:

The app is showing you all the different projects and its samples. Here we only have two projects (depth and snp_density) and its samples.

Now, select all the depth samples and click Run, that should take you to the visualization page:

Now, enter the region you have in the example that comes in the input text box: Chr17:1100000-1200000 and click vizit. The app will render the visualization for that region for the samples we selected previously:

Now, feel free to incorporate your datasets and use the app to browse them.