domain-gfx

Domain graphics library

in development

A live example is available to play with at https://proteinswebteam.github.io/domain_gfx/

Installation:

This library is available on npm, to install it run:

npm install --save domain-gfx

Otherwise, you can use it directly from the unpkg CDN at https://unpkg.com/domain-gfx

Usage:

Using ES modules, or a module bundler

recommended way

This library is available as a plain ES module, you can use it directly by importing it into your code like this:

import DomainGfx from 'domain-gfx';

Using a global

you can insert the script in your page, synchronously or not. It will add the global DomainGfx.

If you are loading it asynchronously, you can listen for the 'domainGfxReady' event on the document, containing the DomainGfx class. e.g.:

document.addEventListener('domainGfxReady', event => {
  const DomainGfx = event.detail;
  // Use the DomainGfx class...
  // ...
});

From the dist folder, use the domain_gfx.es5.js file, for ES5 support, or domain_gfx.es2015.js, for ES2015 support.

API:

Syntax

new DomainGfx(userParameters);

Parameters

• userParameters, object: contains information to render the graphic

  • parent, Element: container element for the graphic

  • data, optional object: data describing the content of the graphic

  • params, optional object: extra parameters, modifying the rendering (not used at the moment) like the scale of the image, or default sizes

Return value

A new instance of DomainGfx

Properties

• DomainGfx.prototype.data getter data(): returns the internal data stored in the instance

• DomainGfx.prototype.data setter data(value): sanitize value, and sets the result as internal data and returns it

Methods

• DomainGfx.prototype.delete(): clean-up logic, removes event listeners and gets rid of pointers to DOM nodes to be able to GC them

Examples:

// DOM container
const parent = document.querySelector('.container');
// Domain graphics data object (see data syntax section)
const data = {
  length: 100,// sequence of length 400
  regions: [// array of sequence objects
    {
      text: 'domain name',
      start: 2,
      end: 40,
      aliStart: 5,// >= start
      aliEnd: 30,// <= end
      display: true,
      startStyle: 'jagged',
      endStyle: 'curved',
      color: 'blue'
      metadata: {// information for tooltip
        database: 'pfam',
        description: 'text about the domain',
        accession: 'PF00000',
        identifier: 'domain X',
      },
    },
  ],
  markups: [// array of markup objects
    {
      lineColour: '#0ff0f0',
      colour: '#bb5b55',
      display: true,
      vAlign: 'top',
      type: 'Pfam predicted active site',
      start: 5,
      headStyle: 'diamond',
      metadata: {
        database: 'pfam',
        description: 'S Pfam predicted active site',
      },
    },
  ],
  motifs: [// array of motif objects
    {
      colour: '#00a500',
      metadata: true,
      database: 'Phobius',
      type: 'sig_p',
      display: true,
      end: 50,
      start: 30
    },
  ],
};

const dg = new DomainGfx({data, parent});

Gotchas and changes from earlier versions:

As compared to pre-existing older libraries such as PfamGraphics

new way to create a graphic

• instead of passing as different parameters, first a node or a string to select a node, then an object with all the data, this now expects an object with a parent, and optional data and params keys.

• instead of calling the render method afterwards on the DomainGfx object, the render is done automatically when instantiating the object, and also everytime different data is assigned to the object.

differences in the data object

Old data objects should still be able to be used, as this library still accounts for the old formats. If it is not able to use the data please upgrade the data object to a newer version.

Should be used:

• American spelling. e.g. colour -> color
• Keys in camelCase. e.g. v_align -> vAlign
• Correct type where need be. e.g. start: integer, display: boolean

Todo:

• Add more documentation
  • shape of data
  • details for developers
• Add more unit tests
• Add TravisCI integration
• Handle parameters (entity/text sizes, x/y scale, etc)
• Automatic color choice for text according to best contrast with background