For years Japan has been trying to reform the way it does science. Since 2001, almost all research institutions, including the entire university system, have been given far greater autonomy in an attempt to make them more competitive with each other and with rivals abroad. Generous funding has also been ploughed into strategic research projects.

The Protein 3000 project, for example, has sought to find the structures of 3,000 proteins in five years, ending next March. These structures will, it is hoped, prove useful in understanding larger proteins that are difficult to analyse directly because of their size. A project called BioBank has also assembled a huge collection of disease-related tissue samples: from June 2003 until this October, it took samples from 162,545 patients — a total of 231,049 disease-related samples, because some patients have more than one disease — on its way to meeting its goal of taking samples from 300,000 patients by March 2008. The samples will be used to search for genetic factors behind 47 diseases, including lung and stomach cancer, diabetes and stroke.

But with the institutional reforms well under way and some of these projects set to wind down, the time is now ripe for Japan to revisit its science policy. In September, Shinzo Abe took over as prime minister and pledged to do just that, announcing plans for Innovation 25, a long-term review of Japan's science and technology needs until 2025. He also took the unprecedented step of appointing a scientist, Kiyoshi Kurokawa, as a special adviser to the cabinet. It is commendable that Abe is taking science policy seriously.

There are a couple of recurrent problems in Japanese science that the new government needs to address. One is an academic system that, notwithstanding the recent reforms, allows élite professors to sit at the top of a rigid hierarchy while junior colleagues patiently await their chance to move up the ladder. More changes are needed to shake up this system.

A second problem concerns the conception and design of major projects such as Protein 3000 and BioBank. These projects have been set up with explicit numerical goals that the public can readily understand. Despite being on target to achieve these stated goals, the government now wonders whether the projects actually best serve Japan's scientific interest. A probable redirection of government support could mean that some of the significant resources that have been invested in these projects — such as the costly machines used to produce protein structures or the samples donated by patients — will go to waste. More meticulous planning and greater consultation with scientists at the planning stage could arguably have assured better use of project funding.

The new science adviser should try to find ways of getting more scientists involved in the development of science policy.

This last issue reflects, in part, project management by civil servants who rotate rapidly through positions every two or three years. They are forced to make important decisions on science policy but are rarely around to be held accountable for outcomes. Japan needs to move more responsibility for the nuts and bolts of such project planning towards advisory boards of scientists who have the necessary expertise and can, at least in theory, be held accountable for the advice that they offer.

One bright spot for science in the new prime minister's record so far is his determination to improve relations with influential neighbours such as South Korea and China. This could — and should — open the way for greater regional collaboration in science and technology.

Kurokawa, meanwhile, has the contacts and experience to lend valuable impetus to Japanese science policy. As a former president of the Science Council of Japan, which represents 790,000 Japanese scientists, he has not been afraid to speak out on scientific issues in the past, and has a broad grasp of the strengths and weaknesses of Japanese science.

The new science adviser should try to find ways of getting more scientists, from both inside and outside Japan, involved in contributing to the development of science policy in Tokyo. With the backing of the prime minister, Kurokawa is well placed to make a difference, and to build on earlier reforms to further increase Japan's scientific productivity.