We wish we do not have to do this, because we are incredibly swamped with genome and transcriptome assembly projects, algorithm development and other work. However, no matter how hard we look around, we cannot find a web-based genome browser that solves our problems. Please talk us out of months of coding by showing an example that we can use. The kids will thank you, when they grow up !!
Here are the features we are looking for in our to-be-favorite genome browser.
1. It should self-install through few clicks in any web folder that we place the zip file in. We do not enjoy installing PERL modules, solving dependencies, and we do not want to see the word ‘cgi-bin’ in year 2013. We did not like cgi-bin even in 1998.
2. It should display genes and genome at various resolutions, and should move very smoothly like Google map. Also, it should do google map-type synchronous move for multiple genomes. Nobody works with a single organism any more.
3. It should have an admin module to set various levels of data access permissions for various users and groups. Some data sets will be public, some will be private for user A and B, some will be private for user C and D. We work with many groups, and even among those groups, we have subgroups with different interests. The bioinformaticians working on computational cleaning up of genes want the latest edition of genes. The biologists digging through data want more stable release.
4. It should have modular code so that anyone can add features easily. Want to add BAM files? Just write a plug-in within the larger code-base. The genome browser does not have to provide APIs, but some kind of documentation on code would be very nice.
5. It should not come with GNU-GPL type license, and God forbid UC style license (please check kind of Browsers to avoid).
6. This is a minor point. We like to change the theme easily.
We could not find a single browser that met 1-5, and we are puzzled why. Outside the bioinformatics world, hundreds of free, open source, high-quality easy-to-install codes exist for everything from secure shopping cart to blogging platform or auction site.
We will start our development, and continue until we are done or a high- quality alternative shows up.
Our design will be highly modular. More than writing a genome browser, we like to design code, whose components can be easily replaced or exchanged. Like the admin module, but not front end? Feel free to write your own jquery, node.js and gRaphael.
Our code will be MIT-licensed. Take it, break it, sell it as a package or do whatever you want to do with it. We will feel grateful that you use our browser. Most likely we will be the only user.
Server code will be PHP-based. We will use a popular MVC framework.
Data side will try to cover all commonly-used databases including MySQL and postgreSQL.
Right from start, we plan to create extensive documentation on the code as well user-functions. We are designing the code to be easily extensible so that anyone can add feature he wants.
This project is to honor sequester. So, we will NOT seek any kind of government funding, not now, not in the future. If you like to be a private sponsor or help us with coding, please feel free to contact at samanta at homolog.us.
This is a side project born out of frustration with status quo. Although we have proof-of-concept code for every feature proposed above, in coding, it takes a long time to turn proof-of-concepts into something presentable. Expect things to move very slow except when it moves fast :) We will continue to work on genome and transcriptome assemblies as our primary project, and come back to this coding only intermittently, unless there is huge community interest to see this project moving.
You can track our progress here at github, and we will write in the blog before any beta release or major release.
Any thoughts or suggestion?