The exclusive club for postdoc exiles

The EU calls them TMRs, cross-border postdoctoral centres that it funds to the tune of Pounds 42,000 each. But, asks Judy Redfearn, how are they faring?

About 2,000 European postdoctoral researchers are working outside their home countries in scientific networks funded by the EU's training and mobility programme (TMR). Although none of the networks is more than half-way through its term, the European Commission was keen to assess their progress. So network coordinators, the postdocs involved, civil servants and a few senior figures in European science met in Graz, Austria, last month to air their views.

Anyone involved in a network belongs to an exclusive club. Only 96 out of 1,4 proposals were funded in the first round - a success rate of less than 7 per cent. The conference consensus that the network, which is part of the current Framework IV research programme, is working pretty well might have been different if some unsuccessful applicants had been present. Nonetheless, the low success rate was seen as a problem, especially as there is little chance of raising it above 15 per cent under the next stage, Framework V.

One solution - to cut the money per network - was rejected because it would mean cutting out the training of postdocs and returning to the type of low-cost coordination-only network funded under Framework III from 1990-94.

"Some of these networks were little more than expensive travel clubs," says Peter Kind, head of the network unit at the European Commission. So taking advantage of the political appeal of training, the EC proposed increasing funding under Framework V to support fewer, larger networks that train one postdoc at each "node" to tackle well-defined, coordinated research projects. The conference did, however, call for greater flexibility over funding to allow more, less expensive networks in non-experimental fields such as mathematics.

The postdocs work at a centre outside their home country and are expected to travel to other centres to experience working in an international team. Each TMR network consists of at least five teams from a minimum of three countries. Seven or eight teams is ideal and 12 is the maximum, the conference agreed.

The most significant selection criterion is scientific excellence, despite the fact that training is the networks' primary goal. "Only good science will provide a stepping stone for a career," explains Alfred Nordheim from the University of Tubingen, who coordinates a network on cell signalling in development and disease involving nine groups from six countries. The pursuit of scientific excellence, however, can be distorted by rules governing, for example, participation by partners from less favoured regions. Despite the scientific community's protests, politics will determine whether such adjustments continue into Framework V.

Proposals with an industrial partner can also win a few bonus points. However, only 69 of the 243 networks funded in the two TMR rounds have at least one industrial node. Most network postdocs see their future in academia: at 30-35 when they finish, they are generally too old or too specialised for posts in industry. "We are anxious to give postdocs an international outlook and open them up to working with industry. The network scheme is a means of enhancing the industrial friendliness of the academic world," Kind says.

Very few small and medium-sized enterprises (SMEs) are involved. But one, Zenon SA, is coordinating a network involving 12 nodes in nine countries to achieve its research goals. The Athens-based company, which manufactures mobile robots for use in health care, built on the contacts established by Nikos Katevas, head of research, when studying for a PhD. "For an SME of 30-50 people, the network is essential," says Katevas. "There is no other way we could afford to do this research".

In the two TMR network rounds, more than 7,600 network partners applied, the vast majority from academia. Their motivation varied. For Michael Rowan-Robinson from Imperial College, London, for example, it was primarily the opportunity to add value to an application for access to data from the Infrared Space Observatory (ISO), a mission of the European Space Agency which ended its active life last month. His network, which is analysing the data, includes most of the leading groups in Europe in extragalactic infrared astronomy.

Money - about Ecus60,000 (Pounds 42,000) per node a year - has also been a strong motivator. But network coordinators also point out the value of forging new, close collaborations. "We had contacts before, though not real collaborations," says Nordheim. "One group in our network uses yeast genetics, another drosophila genetics, two use mouse genetics and one group does crystallography. We are experts from different fields working to a common goal". Competition between the members of the group is not a problem, he says, because their areas of expertise are complementary.

Low selection rates, however, could put researchers off applying for networks. Coordinators at the conference requested greater openness about the selection procedure and expressed concern at the commission's practice of not publishing selection panel members. They also put in a plea for greater flexibility in the way networks are run. "The coordinator should have the power to redistribute money half-way through," summed up Anton Zeilinger from Leopold-Franzens University, Innsbruck who runs a network on the physics of quantum information. That should include the power to eliminate nodes that have not been pulling their weight. Greater flexibility would also allow coordinators to spend money on support staff to take some of the burden of running a network.

There was a plea for support from home institutions to help staff prepare proposals, which is often a lengthy business. Some networks had difficulty recruiting fellows and asked for nodes to be allowed to appoint nationals as a last resort. The commission was encouraged to take a more systematic approach to advertising vacancies.

A survey of 21 network fellows who attended the conference revealed that most are content with the research they are doing, their level of supervision, their interactions with other network teams and their mobility within Europe. But several fellows reported confusion over what their salaries would be, what would be expected of them and what they could expect from their hosts. They called for more advance guidance from the commission.

"The network has provided me with an opportunity to establish contacts I otherwise wouldn't have had," said Denis Le Gourrierec who works at the University of Uppsala in a network on artificial photosynthesis. In common with most others, he thought these contacts would be valuable for the future, although several felt that returning home would be difficult. But a plea for a returner's grant was greeted unenthusiastically by older scientists who had managed their careers without such help.