Professor Rebecca W. Doerge part of $50 million iPlant initiative

06-18-2008

WEST LAFAYETTE, Ind. - The National Science Foundation has awarded $50 million to a team of universities including Purdue to create a global center and computer cyberinfrastructure to address plant biology's grand challenges.

The project, called the iPlant Collaborative, will for the first time unite plant scientists, computer scientists and information scientists from around the world to provide answers to questions of global importance such as food supply and climate change.

Purdue professor of statistics and agronomy Rebecca W. Doerge said the iPlant initiative covers everything from basic science to challenges facing the world.

"The purpose of iPlant is to unite plant sciences on a global level to attack big issues, as well as to advance the fundamental science of how plants work," said Doerge, who also is director of Purdue's Statistical Bioinformatics Center and Purdue's lead investigator for this collaboration. "Purdue was chosen because of our exceptional group of statistical bioinformaticians who have focused on plant science. Statisticians will play a key role in the success of this initiative by helping to design effective experiments and analyze the data."

The five-year project is renewable for a second five years for a total of $100 million. The project is centered at the University of Arizona, and other institutions involved in the project include Cold Spring Harbor Laboratory in New York, Arizona State University and the University of North Carolina at Wilmington.

The iPlant Collaborative will create both a physical center and a virtual computing space where researchers can communicate and work together to share and analyze data. The community will put forth ideas of challenges to be addressed by the collaborative, and teams of experts from around the world will be assembled.

"The iPlant team has a compelling vision for an organization by, for and of the community that will bring to bear the power of cyberinfrastructure to enable scientists everywhere to take on some of the most important questions in plant science," said Joann Roskoski, executive officer of the National Science Foundation Directorate for Biological Sciences.

The collaborative will use new computer, computational science and cyberinfrastructure solutions to address the evolving array of challenges in the plant sciences.

Purdue's statistical bioinformatics group is recognized for developing novel statistical methods to analyze data from new biotechnologies, which will be important for the high-level technologies used to study genomic issues, Doerge said.

"The input of a statistician in the conception and design of an experiment can help ensure that the data, when analyzed properly, will be able to provide significant conclusions," she said. "Statisticians can help guide plant scientists in the design of their experiments, collection of data and through the analysis of their data. Often what is learned from one experiment provides the direction for the next experiment."

The collaborative also focuses on community building and educational enterprise. Students, teachers and the public will have access to iPlant's resources and data, as well as to educational tools designed to help them understand that data and develop inquiry-based learning modules for K?12, undergraduate and graduate science education.

Purdue will contribute to the educational goals of the iPlant Collaborative through its student-run volunteer organization STATCOM.

First created at Purdue and run solely by graduate students, STATCOM provides free statistical consulting to the community. Through informational material and on-site visits, STATCOM has grown and been established at universities throughout the world.

"The goal is to develop a network of statisticians and to grow STATCOM at universities participating in these grand challenges," Doerge said. "We want STATCOM to branch out to more universities and communities big and small. We want to expose the general public to the benefits of statistics and how it can be used to answer questions."

The iPlant Collaborative will train the next generation in computational thinking and is designed to be able to reinvent itself as needs and technologies change.

"This is a great way to expose our students to real science in a very forward-thinking way," Doerge said. "This is their future. Science is continuing to move from the single investigator project to large interdisciplinary projects that require biologists, computer scientists, statisticians, mathematicians and all sorts of disciplines."

The iPlant Collaborative will cull information from existing databases of a variety of plants. Because plants are very interrelated, what is learned from one can often help in the understanding of another, Doerge said. The consolidation of existing databases will allow for analysis across a broad range of species and will provide a wealth of knowledge.

"iPlant will reach out, take the information that is there, identify the missing pieces and then bring experts together to work to fill in the blanks," she said.

One feature of iPlant that will be developed is the ability to map plant biology research in much the way that Google Earth physically maps our planet. Like users of Google Earth, users of iPlant may one day be able to "zoom" in and out among various levels of plant biology, from the molecular to the organismic to the ecosystem level.

For example, a researcher might "zoom in" to analyze the carbon fixed, oxygen produced and water utilized by individual leaves, then "zoom out" to analyze how all of these might affect large-scale changes in ecosystems and how that could affect air quality and climate.

The collaboration also will be used to investigate the sociological aspects of how scientists interact.

"Through the iPlant Collaborative, the NSF has enabled a vast range of scientists to work on complex, global problems. In doing so, the NSF will take advantage of this opportunity to study communications and interaction among scientists," Doerge said. "The idea is to learn what researchers need to function highly together when not physically in the same place and to observe how they talk to each other."

The project is currently in a development stage. In April the first conference will be held at Cold Spring Harbor Laboratory to evaluate grand challenge suggestions.

Writer: Elizabeth Gardner, (765) 494-2081, ekgardner@purdue.edu

Source: Rebecca Doerge, (765) 494-6030, doerge@purdue.edu