##NEW 09-26-2014 - Synapse Annotations. Load annotations from a separate file with format "x y z\n" for each annotation. You can use annotation_parser(annotation_text) to parse it into the format [{'x': x, 'y': y, 'z':z}, {'x': x1, 'y': y1, 'z':z1}, ..., {'x': xn, 'y': yn, 'z':zn}] and load that into the annotation property of SharkViewer.

08-08-2014 - Three.js loader version: html/js/sharkviewer_loader.js If you want to use the neuron rendered from SharkViewer in another scene use this version. html/Shark_Viewer_Loader_Version.html shows an example of how to use it.

06-18-2014 - Serverless SharkViewer! Now the html/Shark_Viewer.html file allows you to load and view your own SWC files without a server. Simply download the repository and open Shark_Viewer.html in a browser, then use the form to view your own SWC files.

##DESCRIPTION SharkViewer is a web-based viewer for SWC neuron files. It is written entirely in JavaScript using Three.js. To view SWC format guidelines visit: http://research.mssm.edu/cnic/swc.html

##EXAMPLE http://janeliascicomp.github.io/SharkViewer/

##QUICK START

1. Open html/Shark_Viewer.html in a modern browser.

2. Click 'Choose File' button to load a SWC file.

HOW TO RUN

1. Load javascript files onto page

<script src="js/threejs/three.js"></script>
<script src="js/threejs/TrackballControls.js"></script>
<script src="js/shark_viewer.js"></script>

2. Load a SWC file into json. There is a helper function swc_parser(swc_text) available, or you can write your own to process it server side and load via AJAX.

var  swc = swc_parser(document.getElementById("swc").text);

3. Create a new SharkViewer object. At a minimum you need to give it the SWC JSON and a DOM element to attach a canvas to.

var s = new SharkViewer({ swc: swc, dom_element: 'container' });

4. Initialize the SharkViewer

s.init();

5. Animate the SharkViewer

s.animate();

Shark_Viewer.html shows an example of this setup.

swc json object:
               { id : {
                       type: <type number of node (string)>,
                       x: <x position of node (float)>,
                       y: <y position of node (float)>,
                       z: <z position of node (float)>,
                       parent: <id number of node's parent (-1 if no parent)>,
                       radius: <radius of node (float)>,
                       }
               }

##OPTIONS swc - swc neuron json object Default: {}

dom_element - html element that will receive webgl canvas Default: 'container'

mode - sphere (full 3-D rendering for nodes and connections, slow), particle (uses particle system to speed rendering and make it look better), or skeleton (no nodes, just connections, very fast) Default: 'particle'

HEIGHT - height of canvas Default: window.innerHeight

WIDTH - width of canvas Default: window.innerWidth

center_node - which node to center neuron on (starts at 1), -1 to center at center of bounding box Default: 1

show_stats - shows fps rendering stats if true Default: false

effect - Renders using effects to allow viewing with specialized equipment: noeffect (none), parallax (parallax barrier), rift (oculus rift) Default: 'noeffect'

show_cones - show cones between cylinders for particle and sphere mode Default: true;

colors - color array for display, nodes of type 0 show as first color, etc. Default: [ 0x00be9e, 0x3919cb, 0x7d0bc4, 0xff6700, 0x3eef00, 0xffce00, 0xf50027, 0x606060, ]

metadata - array of javascript objects with labels and type numbers to display as legend (type number matches to type column in swc file) format: Default: false Example:

[{"label":"undefined","type":0},{"label":"soma","type":1}]

annotation - array of javascript objects with x, y, z coordinates to display as synapse annotations. Default: false Example:

[{'x': x, 'y': y, 'z':z}, {'x': x1, 'y': y1, 'z':z1}, ..., {'x': xn, 'y': yn, 'z':zn}]

annotation_color - color to display annotations Default 0x111111

##Author Information SharkViewer was written by Charlotte Weaver ([email protected]) and Christopher Bruns ([email protected])

Scientific Computing
Janelia Farm Research Campus
Howard Hughes Medical Institute