Modern biomedical research depends on free access to an ever-expanding list of online data repositories such as GenBank and the Protein Data Bank. These increasingly expensive bioinfrastructures need stable and continuous funding — the problem lies in getting agencies to provide the cash.

This is a particular challenge in the European Union (EU), where the natural caretaker of such widely shared infrastructures is the EU's executive body, the European Commission. But the EU has always been adamant that this is not the commission's job — it asserts that the responsibility lies instead with the individual member states.

In an effort to achieve at least some coordination, the commission last August launched the European Research Infrastructure Consortium (ERIC). This legal framework aims to make it easier for a group of EU member states to band together to pay for a particular piece of research infrastructure without getting tied up by their local legal requirements. The first in line for ERIC status is the Biobanking and Biomolecular Resources Research Infrastructure, which has its executive offices at the University of Turku in Finland.

Although the ERIC concept is extremely helpful, it is not without its own dangers. ERIC-fatigue, for example, could easily set in among member states that, having paid for several ERICs, may not want to sign up to any more. And individual countries in a consortium could cause upset years down the line by baulking at necessary expansion plans.

Sitting above its member states' national interests, the commission remains the logical caretaker for EU-wide research infrastructure. And now is the time to change the EU's mind about its role. Negotiations are just beginning on the shape of the Eighth Framework Programme on research (FP8), the EU's next multibillion-euro funding programme, due to launch in 2014 (see page 722). Experts say that it should be possible to have a dedicated budget line for infrastructure in FP8 — if scientists aggressively and systematically help to make the case for it.

This may not come naturally to them. Even more than their US colleagues, European researchers tend to regard the idea of lobbying with distaste. But that attitude is beginning to change. Last year, for example, ten biomedical infrastructures — including Infrafrontier, which phenotypes and archives mouse strains, and the European Clinical Research Infrastructures Network — took the unusual step of hiring a small science-oriented public-relations company in Brussels to help represent their collective interests to the power brokers in the EU, including the parliament and the council. The infrastructure group has not yet pushed the FP8 option specifically, but it should do so — and other researchers should join it.

As the EU's central administration has become more powerful, lobbying activities in Brussels have grown. Lobbyists from other sectors — including the anti-science faction — have no scruples about fighting to win. Those who stay out of the rough-and-tumble will lose.