Image courtesy of Asif Akbar.

iPlant is going to solve the grand challenges of plant biology. It’s a big bite to chew, but the creators of iPlant think big.

“One of the really neat things about this project is that it was funded to exist by, for, and of the community,” said Dan Stanzione, co-director of the iPlant project. “The idea was to be collaborative from square one, and not to take the ‘if you build it’ approach.”

That’s why the developers behind iPlant didn’t propose specific tools or structures. Instead, they created a process whereby the plant biology community could come together and decide what the field’s grand challenges are.

Stanzione estimates that over 100 faculty members from forty different institutions worldwide participated in the grand challenge process. “Most everything was done remotely,” said Stanzione. The tools used included teleconferences, video conferences, collaboration via a wiki site, and email.

Two of the six grand challenges that were posed as a result of this process are currently being developed via iPlant. The first is in the area of phylogenetics. Dubbed the iPlant Tree of Life, iPToL will develop the needed cyberinfrastructure to assemble a comprehensive “tree of life” for the planet’s half million known species of green plants.

“The second is actually a merger of a couple of different proposals that came in,” said Stanzione. This grand challenge is known as “Relating Genotypes to Phenotypes in Complex Environments,” or iPG2P for short. iPG2P will attempt to understand how the genotype of plants influences their phenotypes in a variable environment.

Now, the researchers associated with each grand challenge have divided into working groups exploring specific problems within their grand challenge. At the same time, an engagement team of about seven software developers are assigned to each grand challenge.

“We’re building a set of web-enabled discovery environments which are software platforms that use the language of biology and support the workloads of biology,” said Stanzione.

The first step was to start developing the technological requirements for each grand challenge group. At this stage, the researchers must answer questions such as, “What is your typical work flow?”

“Our team works with them to develop what the software should look like,” said Stanzione. “And from those requirements we start to build the software.”

A number of tools already exist, but the iPlant developers hope to hook the best of existing tools together to create a stable, scalable environment.

It’s a development-intensive process, but that’s part of the point. “There is new software for each one, but as we do more and more of them there will be less new components with each one,” said Stanzione. “And we do look for places to have synergy across the grand challenges.”

As a bonus, the tools developed for iPlant could be applied to other areas of biological research.
One of the big challenges iPlant faces is the integration of data sets. “There’s a huge amount of publicly available data and the number one problem is getting them all to work together,” said Stanzione. “It’s simply a huge amount of work to make it all fit together.”

The actual data analysis will require a great deal of processing power. Some of the problems iPlant will solve are scalable, involving small tasks that are repeated over and over again. Others are tightly coupled, and will require access to a supercomputer to run at reasonable speeds.

The computations and data storage will be handled by TeraGrid and the Texas Advanced Computing Center; Stanzione is the deputy director of TACC.

In the past the largest sustained research efforts have been in the field of particle physics, bringing us the likes of the Large Hadron Collider. But iPlant may be a sign that large science collaborations are for everyone in the information age. “It’s overwhelming to me what a daunting challenge this is,” said Stanzione. “To really make progress, you need a big, sustained effort.”

—Miriam Boon, iSGTW