For many people, the term “computational systems biology” invokes projects like E-Cell in Japan or BioSpice in the United States, but the European Union is taking steps to level the international playing field. Through its Sixth Framework Program, the EU recently funded two software projects to help integrate the fragmented state of systems biology research among EU member states.
Mirroring the goal of systems biology itself, the result is expected to be of greater value than the sum of its parts, said Frederick Marcus, a scientific officer for the European Commission who oversees bioinformatics and computational systems biology projects. “We try to do what the member states can’t,” said Marcus. “It’s a basic requirement [for FP6 funding] that several countries participate with the aim of linking together different national programs and national resources … So these basically act as glue for rather larger efforts at each of the individual labs.”
Funding for the projects is modest by US standards — and even by those of many EU member states — at around €2 million ($2.4 million) per project. The amount, spread over three years and across multiple labs, is just about enough to fund “the equivalent of a few postdocs at each lab,” Marcus said.
But despite the limited funds, the two collaborative projects — EMI-CD (European Modeling Initiative — Combating Complex Diseases) and Combio — are viewed as important steps in building an EU-wide systems biology research community on a par with similar efforts in the US and Asia.
Keeping up with the Joneses (and Kitanos)
Marcus recently coordinated a workshop on European computational systems biology in an effort to help guide future EU funding policies. A report on the workshop issued in March (ftp://ftp.cordis.lu/pub/lifescihealth/docs/csbworkshop_2004_03_en.pdf) outlined several key challenges for European competitiveness in the field, including fragmented research “at multiple locations, with little coordination, and inadequate funding,” a “lack of available funding compared to the US and Japan,” and the fact that “the US and Asia are much more active than Europe in modeling and simulation of complex processes.”
The workshop identified a number of national-level projects in computational systems biology within Europe, but the report pointed out that the disjointed nature of these efforts will require “substantial EU funding tied to structuring the field, in combination with national funding agencies, … to foster computational systems biology in Europe.”
The report recommended near-term efforts such as the STREP (Specific Targeted Research Projects) mechanism that supports EMI-CD and Combio, but conceded that this modest level of funding would prove inadequate in the long term. “As a consequence of the broad applicability and the involved tools, one or two large projects and networks are insufficient to advance computational systems biology in Europe,” the report stated. Future funding efforts should support “strategic projects that provide both conceptual computational and experimental advances together, using relevant model systems.”
Marcus was unable to discuss the EC’s future funding plans for computational systems biology under FP6, and a spokesman for Philippe Busquin, the EU research commissioner, said that it is “too early to go into details” regarding funding priorities for FP7, which is now under discussion.
Hans Westerhoff, head of the department of molecular cell physiology at the Free University of Amsterdam and leader of the Netherlands-based Silicon Cell project, estimated that it will take at least two years for the EU to issue a “major call” for an EU-wide systems biology effort. Westerhoff added that Europe was a bit ahead of the US in systems biology around five years ago on the national level “on the basis of local grants,” but the recent surge of interest — and federal funding — for systems biology in the US has resulted in a “quantum jump” that has left Europe behind.
“Comparing it to the US or Japan, I think the EU will have to increase its funding to give us a competitive position,” said Ralf Herwig, bioinformatics group leader at the Max Planck Institute for Molecular Genetics and principal investigator for EMI-CD. Herwig noted that bioinformatics “as it is now” is quickly evolving into systems biology, which means that EU funding patterns will have to change. “What we really need is integrative approaches,” he said. “There is no need, for example, to develop another clustering algorithm for finding co-regulated genes, because any algorithm will do it. What you really need is to add functional annotations to these clusters.”
Building Systems Biology Bridges
Funding constraints — coupled with a plethora of existing efforts — forced the FP6 project evaluators to carefully consider the best way to begin building an EU-wide computational systems biology infrastructure. Current national-level systems biology activities ranged from basic data analysis, to predictive modeling of simple systems, to whole-cell simulation, Marcus said, making the fundamental question “finding a match between the data and what you’re trying to do.” In some cases, Marcus added, the modeling was “running ahead of the data to support it” — a situation that the EU did not want to encourage.
Marcus said that EMI-CD and Combio were funded, in part, because they have set out to “match the level of modeling to the level of knowledge to make it work.”
In the case of EMI-CD, the goal is to develop a software prototype by the summer of 2005 that will allow users to simulate biochemical processes using their own experimental data. The first step in achieving this goal, according to Herwig, is data integration. EMI-CD, he said, will address the “gray area” that currently exists between “low-level” large-scale data sets and “high-level” kinetic models.
In addition to Herwig’s group at MPI, EMI-CD consortium members include Lion Bioscience, MicroDiscovery, the European Bioinformatics Institute, and Tel Aviv University. The system will be built on a modeling and simulation platform currently under development at MPI, but Lion will contribute its SRS database integration system, MicroDiscovery will provide software for integrating experimental data, EBI will offer access to its wealth of databases, and Tel Aviv University will contribute its “know how” in genetic network analysis, Herwig said.
Combio is a collaboration among the European Molecular Biology Laboratory, Spain’s National Center for Biotechnology and National Cancer Center, Israel’s Weizmann Institute; the Free University of Brussels, Germany’s University of Göttingen, Budapest University of Technology and Economics, and Biobase. According to the EC Computational Systems Biology workshop report, the partners plan to test a number of current modeling and simulation tools with the goal of developing a set of guidelines that specify which simulation methods are best suited to specific research problems, given a certain amount of data. The consortium will focus its efforts on simulation methods for the p53 regulatory network and the spindle formation during mitosis, the report said, and will draw from the aMaze and Transpath databases.
Participants in the Combio project were not available for further comment.