myExperiment: A social network for workflows
January 10, 2008
I have more backlog than I can think about, but slowly but surely, I am going to start working through all the posts that I started and never finished towards the end of last year. The first one I want I want to finish up is about project that I first heard about at Scifoo. There, I had a chance to meet Carole Goble and Duncan Hull. Carole is the Director of the myGrid project, which also includes Taverna and myExperiment, and Duncan counts myExperiment as one of his current projects.
So what is myExperiment. At Scifoo, Carole referred to myExperiment as a “Facebook for sharing workflows”, which is how Duncan describes it as well. More formally, myExperiment is described as follows
The myExperiment Virtual Research Environment enables you and your colleagues to share digital items associated with your research — in particular it enables you to share and execute scientific workflows.
The myExperiment Wiki uses the following description
myExperiment makes it really easy for the next generation of scientists to contribute to a pool of scientific workflows, build communities and form relationships. It enables scientists to share, reuse and repurpose workflows and reduce time-to-experiment, share expertise and avoid reinvention.
Workflows here come primarily from Taverna, but in theory all workflow systems are supported. The figure below shows you the myExperiment idea graphically
Essentially what you have is a community around scientific workflows. What makes this concept especially powerful is the fact workflows are a key aspect of science, and that there could be multiple workflows tackling the same problem. If I think back to my past, something as simple as the way one prepares a protein for a MD simulation, captured as a CHARMM script can vary from one scientists to another, even if broadly they are doing the same thing. What myExperiment allows you to do is learn from the collective wisdom of your peers. In addition, workflows do a wonderful job of digitally capturing the entire “methods” section of a paper, allowing you to reproduce results, capture best practices, etc. An open community like myExperiment fits in beautifully with the open science/open data world. In fact, I hope there comes a day when scientists capture their protocols (for computational projects) and deposit them in myExperiment or similar resource. Along with open data access, being able to reproduce protocols is also critical.
I should add that Carole, Duncan and their colleagues, absolutely get the web. Not only do they understand the scientific potential, but they also understand how the web can be used as a marketplace (a powerpoint presentation on the wiki uses that specific term), as a gateway, and as a platform for building community. The site is beautifully designed (how many scientists know about the Silk icon set), the entire source code is available (so you or I can set up our own myExperiment instance), they have a simple RESTful API and amongst other things have tested the site against S3. Of course, you have to root for anyone who has the following answer to the question “Why not use a portal solution?”
The first principal of the myExperiment design is “bring myExperiment to the user”. We have deliberately aimed it at the next generation of scientists. Say no to 1970s library interfaces! Seriously, we hope that myExperiment encourages and informs a debate about portal approaches.
Last but certainly not the least, the site supports OpenID. Did I mention that the myExperiment folk get the web
Hopefully they will soon support OAuth, if they don’t already.
I have a vested interest in myExperiment. While I am way behind in my efforts, partly due to the fact that the hardware setup I had acquired is going to be replaced (yes I am finally going Mac), and partly due to work, but one of my goals is to use workflow systems to set up structural bioinformatics/cheminformatics workflows using web standards and to see if I can incorporate Amazon’s web services and resources like Freebase. If I get there by summer, will be very happy, since work is going to keep me really busy.
Technorati Tags: myExperiment, myGrid, Carole Goble, Duncan Hull, Bioinformatics, Workflows
Add New Comment
Viewing 3 Comments
Thanks. Your comment is awaiting approval by a moderator.
Do you already have an account? Log in and claim this comment.
Do you already have an account? Log in and claim this comment.
Do you already have an account? Log in and claim this comment.
Do you already have an account? Log in and claim this comment.
Add New Comment
Trackbacks