Greetings from the iPlant Collaborative! With this e-newsletter, iPlant begins its third year. Read on for more information about our plans for “iPlant 2010 Conference” in Las Vegas later this spring. Also in this issue, we introduce a regular column by iPlant’s Semantic Web Architect, Damian Gessler, who’ll explain what the Semantic Web is and why you should care, as well as other topics about the future direction of the World Wide Web. As always, we welcome any comments or suggestions for our newsletter; please send to feedback@iplantcollaborative.org.
To unsubscribe, please click here (then hit ‘send’, nothing else is required). If this email does not display correctly, copy and paste this link into your browser to view The Leaflet on our website.
What is the Semantic Web and Why Should You Care?
By Damian Gessler, iPlant Semantic Web Architect, dgessler@iplantcollaborative.org.
 |
Currently, the Internet is connected by an intricate web of links. Semantic Web is an effort to fundamentally alter the way these relate to each other.
Photo Courtesy of The Opte Project |
The Semantic Web is the linking of concepts and statements on the World Wide Web in a manner discernable for contextualized decision-making by machines. To understand the Semantic Web today and its relevancy to iPlant and science, we need to go back and look for the signatures and commonalities of its intellectual antecedents.
In the early 1990s, computing was still a rather traditional, structured endeavor. Yet transparent to the user, underneath lay a creative and non-traditional foundation. A quarter century earlier the research arm of the U.S. Department of Defense led an effort to construct an informatics and communications network that could withstand unpredictable service interruptions while maintaining connectivity across a broad swath of sub-networks, protocols, and operating systems. The designers were sensitive to the well-known issues with circuit-based communications systems—the telephone system being the canonical example.
Instead of addressing this problem with hardware, such as adding more circuits or increasing circuit redundancy and reliability, information architects such as Lawrence Roberts, Leonard Kleinrock, Robert Kahn, and Vinton Cerf struck a bold path: they diminished the role of the circuit to being simply an infrastructural artifact and moved the “smarts” of routing and marshalling data from the circuit system level to the level of operations on metadata riding with the data itself. Under this new model, data was broken into “packets” each containing a copy of its destination address. Nodes in the network would look at incoming packets and determine the best routes and hops based on local conditions.
We know this system today as the Internet—brilliant invention, no longer completely understood by any one person, truly bigger than the sum of its parts. The decision to move the routing away from a top-down systems level to a myopic, information-centric model was by no means obvious and had its share of detractors. Yet this move allowed a true network to evolve and scale.
Meanwhile, hypertext—the linking of content from one document to that of another—was maturing under the thinking of Vannevar Bush, Ted Nelson, Douglas Engelbert, and others. But it was not until the early 1990s that these two inventions—the Internet and hypertext—came together in what we now call the World Wide Web. The World Wide Web is a document-centric, hypertext-linked network of content riding on the Internet. The fabulously successful World Wide Web is the work of many, but like the Internet, the brainchild of a few (Sir Tim Berners-Lee along with Robert Cailliau). It is architecturally based on the perspective that a network of components (e.g., documents) creates a value greater than the sum of its parts.
Both the Internet and the World Wide Web were paradigm-changing because they enabled a larger usage: they solved a problem in a way that enabled the very use of the system to generate a greater, synergistic value. Building on this intellectual foundation, the Semantic Web takes this approach to its next logical extension.
Connecting computers, like dumb machines exchanging bits, gets one only “so far.” Interoperability is not integration. Nor does mere aggregation of data generate synergistic value in knowledge. Designers of the Semantic Web realize that knowledge comes not from just words but from connecting contextualized information. Humans do this every day, but for computers, discerning meaning so as to determine suitability-for-purpose requires computable semantics. Some might think this necessitates some artificial intelligence more appropriate for science fiction than science practice, but the history of modern informatics—the Internet and the World Wide Web—hints that this is not so. Like the Internet and the World Wide Web, the way this is being done in the Semantic Web community is non-obvious, highly infrastructural, and potentially game-changing. The Semantic Web has application to iPlant because, in building a cyberinfrastructure for plant scientists, iPlant has a constituency deep in distributed, interdependent, yet un-integrated data and knowledge. The Semantic Web enables an integration of those data and services driven by the decentralized creative contribution of others; its intellectual antecedents hint that the Semantic Web has the right ‘pedigree’ for the challenges we face.
Damian Gessler will be writing a regular column on Semantic Web issues in The Leaflet. If you have questions or comments about the Semantic Web that you’d like him to consider in future columns, contact him dgessler@iplantcollaborative.org
iPlant Action Teams: Grand Challenges in a Microcosm
By Ann Stapleton, iPlant Faculty, stapletona@uncw.edu
An iPlant Action Team (iPAT) consists of a computational faculty member and a plant biology faculty member collaborating and developing a common vocabulary as they solve a mini-project in computational plant biology with support from iPlant personnel.The first iPATs are making progress toward their goals, which include developing tools for trait reconciliation using genome structure with the iPlant Tree of Life (iPToL) Grand Challenge project, and developing analysis tools for short-read sequence data for educational use with the iPlant Genotype-to-Phenotype (iPG2P) Grand Challenge project and Dolan DNA Learning Center at Cold Spring Harbor Laboratory (CSHL).
The California State University Long Beach team, comprising computer scientist Min He and plant biologist Judy Brusslan, is working with short-read Nextgen sequence analysis tools. Judy comments that "Min and I are enthusiastic about developing a course that utilizes short- read sequence datasets for undergraduate bioinformatics projects. There are no bioinformatics faculty in either of our departments at CSULB, so this iPAT has provided a framework to initiate our analysis. Our biggest challenge is that we often feel like we speak two different languages; it takes time to be sure the other person understands what you’re trying to say."
 |
 |
| Vic Alberts |
David Sankoff |
The second iPAT, comprising plant evolutionary biologist Vic Alberts from the University of Buffalo and computational mathematician David Sankoff from the University of Ottawa, is currently refining their project to incorporate specific timelines and details of the gene order evolutionary algorithms to be included in their module. They will also learn how the iPlant development team does needs and requirements analysis and incorporate iPlant development methods into their charter.
What have been the “lessons learned” so far? It is apparent that even small-scale projects being undertaken by the iPATs take time to complete, as team members cannot work on their projects full-time and developing a common language in order to communicate is the first barrier that must be overcome. Long-lasting partnerships like these, with catalysis from iPlant, will allow the collaborators to apply their new approaches to future work and in the classroom.
Modeling Workshop for College Educators
iPlant is partnering with Shodor, the National Computational Science Institute, and North Carolina Agricultural and Technical State University (NC A&T) to offer its second "Computational Biology for Biology Educators" workshop. The focus will be on modeling, with an emphasis on plant-related examples such as the effect of CO2 concentration and light intensity/quality on photosynthesis, models of gene evolution, and ecophysiological models of plant flowering time. The workshop will take place on the NC A&T’s campus the week of July 18th. Watch for a detailed workshop schedule and application information on the iPlant website (www.iplantcollaborative.org) in early March.
|
Alberto MaciasDuarte, iPlant graduate fellow, has accepted a position as a data analyst with the Rocky Mountain Bird Observatory beginning in June. Alberto has been collaborating with iPlant Teacher Fellows over the past two years, supporting implementation of lessons and modules in high school classrooms, presenting at conferences and assisting in the development of modules with a computational focus on plants. Congratulations Alberto!