On the Human Condition
Volume XXVIII Number 2
A "map" of major areas of knowledge in aging research. The height of each peak is proportional to the number of documents in the peak. Labels show the two most common words in the titles of the documents related to the peak.
Rx: Information Collaboration
What a piece of work the human genome is: A microscopic yet expansive galaxy of complex biological and chemical processes that determine a person's eye and hair color, the contours of the face and body shape, perhaps even personality. The genome's frontiers have barely been touched by researchers probing the mysteries of genetic diseases and searching for cures and new therapies.
In 2000, only a few months after the federal government announced that scientists had completed a working draft of the human genome, Indiana University charted a course to become a leader in life sciences research. Fueled by a $105 million grant from the Indianapolis-based Lilly Endowment Inc., the Indiana Genomics Initiative was established, an ambitious plan that taps into IU's expertise in a variety of scientific disciplines.
Most of IU's life sciences research efforts draw upon its investments in information technology such as supercom-puting, and its links to Internet2 and high-speed computing networks such as TeraGrid. Researchers at the IU School of Informatics--the first and largest school of its kind in the nation--also fill vital roles in the university's commitment to life sciences.
"With the explosion of data that has resulted from the deciphering of the human genome and other advances, informatics has embraced the challenge of taking that data and transforming it into information that solves key problems in genetics and medicine," says School of Informatics Dean J. Michael Dunn, who is also Oscar R. Ewing Professor of Philosophy at IU Bloomington. "Collaboration among all scientific disciplines is essential, and the role of informatics will become increasingly important in future life sciences research at IU."
Dream Tools for Diagnostics
As Katy Börner powers up a computer at the IU School of Library and Information Science, complex, colorful visualizations spiral into view on three monitors, displaying information from various scientific disciplines and papers related to melanoma research. Börner isn't a physician but her work is enabling improved patient care and further biomedical investigations into the mechanisms and dynamics of disease.
She and other members of her information visualization laboratory use multiple servers and 10 terabytes (10 trillion bytes) of disk space to store, integrate, and analyze large-scale datasets. Their goal is to enable the medical community to better utilize existing knowledge and data by providing researchers and practitioners with a global map and other tools to identify and make accessible major research areas, experts, institutions, regions, nations, grants, publications, and journals.
"Researchers can quickly gain an overview of existing research, major research frontiers, and trends," says Börner, associate professor in the School of Library and Information Science, who holds an adjunct appointment in the School of Informatics. "Relationships among different data types become obvious."
In another project, Börner contributes her expertise in data analysis and information visualization to help IU School of Medicine researchers improve the treatment of adults and children with cancer and heart disease. In this project, Börner is part of a team of experts in genetics, evidence-based medicine, proteomics, database integration, and mathematical modeling. The group is led by Susan Ragg, an associate professor of pediatrics and an oncologist at Riley Hospital for Children in Indianapolis. Collectively, she and her colleagues make up the IU Center of Excellence in Computational Diagnostics, funded by a $2 million grant from Indiana's 21st Century Research and Technology Fund.
"In most of our projects, the main measures of success are improved tools for knowledge access, management, and education as well as the number of peer-reviewed and cited publications," says Börner. "In this project, the main measure of success will be how many people we can help survive."
Börner also is the principal investigator of a $1.1 million National Science Foundation grant to develop the information technology needed to study large-scale networks. When completed, Network Workbench (NWB) will provide a one-stop, online data-code-computing portal for the scientific research community. The NWB will offer wider access to algorithms that biologists and other scientists can use to carry out network analyses, modeling, and visualization projects in their own fields.
"Biologists will use Network Workbench to analyze and consolidate data from diverse networks in order to understand genes and proteins and the roles they play in cancer and Alzheimer's and Parkinson's diseases and develop new drugs to treat them," Börner says.
Joining Börner on the NWB project are co-investigators Alessandro Vespignani, professor of informatics at IUB; Santiago Schnell, assistant professor of informatics at IUB; Stanley Wasserman, professor in the departments of psychology and sociology at IUB; Eric Wernert, associate director of research and academic computing for IU's University Information Technology Services; and Albert-László Barabási, professor of physics at the University of Notre Dame. Craig Stewart, associate vice president of research and academic computing at IU, is a senior collaborator on the project.
"We have a unique team of experts, many of whom have made major theoretical contributions, and they are now taking on the challenge of building a dream tool for their respective research communities," says Börner.
Network Workbench may prove to be a valuable tool to identify and track emerging diseases around the world, says Vespignani, who is internationally known for the statistical analysis and computer modeling of epidemics and how they spread. In recent, related research, Vespignani led an informatics team that built a first-of-its-kind model to help better predict the outbreak of pandemic influenzas (such as the avian flu) and other diseases that spread along the world's commercial airline routes.
The team used the massive passenger-flow databases of the International Air Transport Association, an organization of 265 airlines that comprises 99 percent of all international air traffic. Census information derived from 3,100 urban areas in 220 countries and related disease patterns from those areas also was collected. Using advanced computation, the researchers were able to run more than 10,000 mathematical modeling equations simultaneously to obtain predictions and confidence intervals.
"From our study, it becomes apparent that the air-transportation-network properties are responsible for the global pattern of diseases," says Vespignani. "The complex features of the network are the origin of the unrelated and seemingly erratic spreading of diseases such as severe acute respiratory syndrome."
"Tools developed by this modeling offer a wealth of information useful in risk assessment and the evaluation of disease-containment policies," Vespignani adds. School of Informatics researchers are beginning to evaluate databases of highway and rail traffic and their interrelations with air transport to further develop disease forecasting models.
Better Living through Chemistry
Medical researchers winnow through and analyze massive amounts of data to find ways to treat disease. Geoffrey C. Fox, professor of informatics and computer science, leads a group of investigators whose work will help accelerate drug discoveries and move them from the research bench to patients' bedsides.
Using a $500,000 grant from the National Institutes of Health, the School of Informatics has established the Chemical Informatics and Cyberinfrastructure Collaboratory. Chemical informatics applies computer technology to chemistry in all of its manifestations, particularly in the pharmaceuticals industry. The collaboratory brings together experts in informatics, computer science, medicine, biology, and chemistry with specialists from IU's Pervasive Technology Labs.
"The tools and infrastructure we are developing will be deployed in real-world industries and academic institutions and will be tested by practicing scientists," says Fox, who also is director of PTL's Community Grids Lab. "Ultimately, our work may help chemists better understand the mechanisms applicable to new methods of drug synthesis and lead to new therapies for cancer, Alzheimer's disease, and other devastating disorders."
The collaborators are devising an integrated cyber-infrastructure composed of diverse and easily expandable databases, simulation engines, and discovery tools, such as PubChem, the NIH's small-molecule chemical and biological database. They also are using new high-capacity computer networks and data repositories as well as developing grid and Web technology for chemistry research.
Other researchers involved in the CICC include Gary Wiggins, director of the chemical informatics program; Mu-Hyun "Mookie" Baik, assistant professor of informatics and chemistry; Peter Cherbas, director of the IU Center for Genomics and Bioinformatics; Keith Dunker, director of the IU Center for Computational Biology and Bioinformatics; Dennis Gannon, professor of computer science; Marlon Pierce, PTL research associate; Beth Plale, assistant professor of computer science; and David Wild, assistant professor of informatics.
Wild is building a prototype for the CICC's Web component and an intelligent agent-based system for potential use in the pharmaceutical industry. Such tools will enable scientists to more quickly gather the information they need to make decisions about which chemical compounds are most likely to be safe and effective drugs.
"The problem today is that early-stage drug discovery is burdened with information overload," says Wild, whose work is funded by a grant from Microsoft. Indeed, overload may be a generous term; the Chemical Abstracts Service, the world's largest repository of chemical and related scientific information, estimates that more than 700,000 new compounds are added to its database each year.
"The prototype we are building will make the computer do the grunt work of sorting the information for scientists," Wild says.
If the 20th century was an age of biology, as some have suggested, the century now unfurling may well be the era of informatics--a field that sorts and arranges the components of the genetic puzzle and helps us understand more clearly what a remarkable piece of work humankind really is.
Joe Stuteville is media relations coordinator for the IU School of Informatics in Bloomington.