A new alliance between the Mayo Clinic and the University of Illinois at Urbana-Champaign could result in a number of new computational tools and other technologies of interest for individualized medicine.
The alliance, announced last week, will provide a framework for broad cooperation between the two organizations in individualized medicine by integrating efforts in three areas: basic, translational, and clinical research; bioengineering, especially for point-of-care diagnostics; and the development of tools and methods in computational biology and medicine.
The partners will work on a number of projects in bioinformatics and other computational sciences, and effort is expected to encompass imaging, nanotechnology, and tissue engineering as well as projects in genomics and the microbiome.
In the past, Mayo has formed similar clinically oriented partnerships with other organizations. Last year, it partnered with bioinformatics company Geospiza to develop software systems that could detect and visualize mutations in sequences that are linked to cancer. The long-term goal of the project, for which Geospiza received $110,000 grant from the National Human Genome Research Institute, is to develop non-invasive methods of detecting cancer. (BioInform 10/09/2009)
Lawrence Schook, director of the division of biomedical sciences at UI, told BioInform that UI would provide high-performance computing for the partnership as well as develop “new computer algorithms to address a lot of the emerging aspects of individualized medicine and computational medicine.”
As an example of the kind of infrastructure available at UI, Schook noted that the university is the home of the National Center for Supercomputing Applications, which will host the 200,000-core petascale Blue Waters supercomputer that’s being built by IBM. The project is supported by a $208 million grant from the National Science Foundation and is expected to come online in 2011.
He said that prior to forming the partnership, both institutions spent a great deal of time examining points of intersection in their research capabilities and interests.
Eric Wieben, a professor of biochemistry and molecular biology at Mayo, concurred. “This is a situation where we are bringing together two partners that have significant overlap in capabilities in the area of biomedical research but each have unique strengths as well,” he told BioInform. “What we will be bringing to the table is some strength in genomics and molecular diagnostics [as well as] as lot of capability in imaging and tissue engineering.”
According to Wieben, Mayo is primarily interested in bringing the concept of individualized medicine to routine clinical implementation. He said that Mayo hopes the partnership will yield biomarkers and pathways to new drug development as well as tools geared for point-of-care diagnostics and other patient-oriented genomics tools.
Wieben said that although Mayo is well known for its clinical care, the institution also participates in a significant amount of research and has a $300 million yearly research budget. “Mayo is not just bringing patient samples to the table,” he said. “We are bringing a lot of expertise in how to do modern patient-oriented research.”
According to Schook, the fact that UI has the IT infrastructure in place has implications for a long-term collaboration between both institutions.
“There are issues that can’t be resolved today with existing technologies and applications,” he said. “We believe that there is a need for the development of new approaches and new educational paradigms.”
He added that the intent of the partnership "is to build the foundations for a longer-term relationship in terms of not only creating the technology but also the human resources required.”
[ pagebreak ]
Building the Right Tools
Rob Rutenbar, a professor of engineering and the head of the computer science department at UI, told BioInform that one of the main goals of UI’s discussions with Mayo is to “identify the gap between the scale of the problems that we foresee coming on the diagnostic side of the world and the kinds of tools that are currently available.”
Although the core expertise is there, Victor Jongeneel, director of bioinformatics at the Institute for Genomic Biology at UI, told BioInform that the infrastructure at the university, particularly for handling large-scale data, was developed for business applications, scientific data mining, and other types of research applications and not specifically for clinical use.
“We have to see what the exact structure of [Mayo’s] datasets are and [then] look at the best way to leverage existing tools and approaches to extract information from them,” he said. “Most of what we are envisioning will be really building new tools based on expertise at Illinois and clearly defined needs and existing datasets at Mayo.”
Jongeneel did point out that there are some tools at NCSA that could be modified for clinical use as well as some statistical methods for classifying diseases based on large-scale expression data.
Rutenbar said he expects that UI will have to significantly reconfigure its tools but added that that’s one of the benefits of working with Mayo.
“What’s interesting for us are these sorts of visions of the future [when] we have access to Mayo Clinic’s vast numbers of patients, vast numbers of cohorts, vast [amounts] of data,” he said. “We can start thinking about asking what the patterns are, what we can see, and what can we deduce that would be relevant to this particular patient.”
In terms of what’s currently available, Rutenbar said that UI has several data-mining tools and brings to the partnership a team that is well versed in looking for patterns in large amounts of structured and unstructured datasets as well as in text and images.
He continued. “We are excited because the Mayo folks are an outstanding deep group of folks on basically the data side and they have clear visions of what they would like to be able to do,” he said. “The gap that Illinois basically fills is that sort of IT computer science, computation, data mining part.”
Schook said that the partnership won’t be restricted to Mayo and UI but will eventually involve additional partners.
“We have already been in discussions with several multinationals with regards to opportunities, as well as commercialization both at Mayo and Illinois in terms of intellectual partners.” he said.
Schook did not disclose any of the groups that the alliance is holding discussions with.
For its next steps, the alliance plans to fund several short-term joint group projects and to that end has issued a request for applications from researchers at both institutions.
Project applications are due in mid July and funding announcements will be made in October.
The pilot projects are expected to last anywhere from six months to a year, after which it is expected that they will have matured enough to qualify for outside funding. Both Schook and Wieben said they expect that partnering together will make them stronger candidates for National Institutes of Health funding and other grant opportunities.
The alliance also plans to commercialize the products of the partnership at some point in the future, although it has not disclosed a specific timeline for when that would happen.