Melding of Minds Results in Sophisticated 'Proteomics' Analysis Methods and Tools
Published June 2007
Several years ago, a number of UC scientists with diverse backgrounds in physics, chemistry, mathematics, computer science and biostatistics came together to establish Cincinnati as a major player in an emerging field known as bioinformatics.
This relatively new field of science merges biology, computer science and information technology so researchers can collect and organize data on how genes encode proteins to function in the body, and understand how that information might translate into new targets for drug development.
Three of the scientists leading those efforts—Jarek Meller, PhD, Mario Medvedovic, PhD, and Alexey Porollo, PhD—are developing a series of expanded Web-based software applications and servers that scientists can use, free of charge, to extract and extrapolate the specific bioinformatic and functional genomic data they need for various research projects.
“Bioinformatics is about managing, but mostly analyzing, large amounts of data coming from many sources—such as sequencing projects, microarray gene expression studies and proteomic profiling,” explains Meller, associate professor of environmental health and a bioinformatics expert. “Our challenge is developing new ways for this data to be manipulated and then transformed into knowledge that will help individual researchers make sense of all the available information.”
Part of these efforts involves developing public domain software applications and Web-based servers that make these new methods available to the scientific community.
The problem, as Medvedovic explains, is that there are literally millions of data points that can give researchers important molecular clues for advancing their research. But there is also a great deal of random variation to sort through to get to that critical information.
“In order to make some kind of meaningful conclusion,” he explains, “you need to establish statistical models that will tell you something about the relationships between the different proteins or genes being studied while accounting for random fluctuations in data.”
“We’re not only using the tools,” adds Porollo, research assistant professor, “we’re making them better and developing new methods for analyzing the data. A growing base of users, and cross-linking by major proteomic resources, such as the Protein Data Bank and various prediction meta-servers, are evidences of the competitiveness of our tools.”
One advanced computer server UC bioinformatics experts are continuing to improve is POLYVIEW, which enables scientists to visualize and analyze the three-dimensional structure of proteins—both those for which molecular data has already been documented and those new molecules that are currently under investigation.
Being able to visualize a structure is a critical step in identifying the functional “hot spots” where scientists can attempt to design small molecules—future drugs—that would intervene with the function of that gene.
Another UC team recently launched a new server, SPPIDER (Solvent Accessibility Based Protein-Protein Interface iDEntification and Recognition), which allows researchers to predict protein interaction sites. Scientists can submit data about a new structure they are studying and, based on complex computational algorithms, make predictions about how it will function in the body. They can also cross-reference this structure-based analysis with other sources of information, such as gene expression patterns, being studied by Medvedovic and his group..
“This will be an important tool for researchers who want to find out which cofactors may impact the behavior of the proteins they are studying,” says Meller, “and we’re excited about making it available to the scientific community.”
Both POLYVIEW and SPPIDER were made possible by grants from the National Institutes of Health and Ohio Board of Regents and support from UC’s environmental health department and the Cincinnati Children’s Hospital Research Foundation. The interdisciplinary effort involves faculty and students from UC’s departments of environmental health, biomedical engineering and pediatrics.
The team estimates that at least 10,000 researchers from 60 countries have used their bioinformatics servers, and more than 150,000 information queries have been submitted since 2004.
For more information on these bioinformatics tools, visit http://folding.cchmc.org/servers.html.