With the near-completion of the Human Genome Project (along with the completed sequencing of the genomes of more than 150 other organisms), and with the availability of systematic data on cellular biochemical networks, it is often claimed that biology has entered the post-genomic era. What does this statement mean? Can the immense flood of genomic information now available revolutionise our quest for self-knowledge and open new discussions among disciplines? Or, on the contrary, will genome projects start an avalanche of new sub-specialties contributing to the fragmentation of science?

Although one will get no clear answer to these questions from the Nature Encyclopedia of the Human Genome , the five massive volumes form the first major reference work aimed at summarising the ramifications of human genome research and its possible philosophical, ethical and commercial implications. The encyclopedia, which has been written by 1,400 experts, includes more than 1,000 articles over about 5,000 pages, with 16,000 references to sources and 1,500 excellent illustrations. It has a comprehensive subject index and a useful glossary. A rough idea of the grandiose conception and execution of the project will be clear from the fact that 11 specialist editors and a distinguished 52-member editorial advisory board shaped the encyclopedia's scope and content.

They interpreted their task very broadly. The concise and mostly well-written papers include not only structural, functional and comparative aspects of genomics, bioinformatics and evolution, but also human diseases and ethical, legal, social and historical issues. Biographies of prominent researchers are also provided, and no one appears to have been overlooked.

There are also many seemingly surprising and even bizarre entries, including "Absolute pitch", "Dolly and Polly", "Nazi scientists" and "Deaf community and genetics". Even social constructivists and feminists are found to be related to the human genome. We found the case studies on genetic counselling, genetic screening and industrial applications especially interesting. Although opinions on these issues often vary remarkably, the articles here strike a balance between naive enthusiasm and ideological contempt.

ÁñÁ«ÊÓƵ

It is understandable why the canvas is so vast. The flood of biochemical information from genome projects has not only provided a veritable blooming of new biological disciplines (such as bioinformatics and comparative genomics), but has also provided empirical evidence in long-standing debates in linguistics, psychology and even philosophy. For the first time, researchers are able to investigate the possible genetic makeup of the human language faculty, or of aggression and altruism, with potential implications for innate qualities in the human mind.

Many of these issues call for a multidisciplinary approach. As emphasised in the preface by David Cooper, the general editor, the major aims of the encyclopedia are to "counter the trend towards fragmentation by promoting an integrated approach... and emphasise the interplay between genetic concepts and principles on the one hand and information acquisition and interpretation on the other". The hope is that the intelligent and concise articles will encourage scientists to see the wood as well as the trees.

ÁñÁ«ÊÓƵ

In many cases, the articles succeed in translating the advances of a fast-developing area into language accessible to non-specialists. With this purpose in mind, articles have labels to tell readers what background knowledge they require. For many subjects there are multiple articles (there are four on the inherited disorder cystic fibrosis) that range from broad overviews suitable for undergraduates to highly technical articles for sub-specialists.

But such a strategy is not always feasible. Many papers for advanced readers cover important concepts (such as evolutionary conflicts) that have little impact on current trends in genome research. More important, the strategy can generate substantial overlap between articles. The dual or multiple papers on the genetics of alcoholisms, artificial chromosomes or the calculation of evolutionary distances offer examples that could easily be multiplied. In the most extreme case of overlap, two short articles on evolution consider opposite sides of the same coin: the selectionist and the neutralist points of view. But the two papers, which are written by the same author, differ from each other only slightly. The encyclopedia should probably have adopted an established style of scientific review journals, with a single longish paper and all specialist details in boxes.

As the general editor readily admits, some articles present personal views on possibly controversial topics. We agree that at this initial stage of genomics, in which there is no consensus on many conceptual issues, it is important to publish differing viewpoints. It was especially enjoyable to read the "memoirs" on sequencing the human genome written by the two rivals, Celera Genomics and the international consortium. Less entertaining is to realise that some experts believe they are the only people worth reading or that they despise their opponents' views. One feels especially doubtful that all the important opinions on a controversial topic have been consulted in articles in which the overwhelming majority of references are citations of the author's own work.

The preface emphasises that "we have deliberately chosen to commission articles from a wide range of authors with contrasting views in the hope that it will stimulate and even provoke further debates". Although each paper is accompanied by a list of related papers, we are sceptical whether the encyclopedia can fulfil the task of stimulating debate. Because papers appear in alphabetical order based on their titles, related topics are often found in different volumes, and hence contrasting views can get lost in the stream of other papers. It might have been more appropriate to group the papers by subject area and to provide a short, balanced introduction by an area specialist editor.

ÁñÁ«ÊÓƵ

But we recognise that there is a valid argument for including some entries more than once, as conceptual categories are not mutually exclusive. Human cloning, for instance, can be considered from various points of view: that of reproductive biology, human genetics, social impact, philosophy and bioethics.

The obvious way to grapple with a fast-moving subject is via the internet.

Indeed, the encyclopedia is intended to be an up-to-date reference work and a dynamic and tractable database with multiple categorisations. This will allow the first edition to remain at the forefront of human genome research. For, as recognised by David Goodman, "the rationale for publishing in printed format may be questioned, since the immense effort yields a reference work with a useful lifespan of perhaps two years". An internet or CD edition could potentially offer network-based knowledge, where details for specialists and related subjects would be only one click from each other. This is in sharp contrast to the linear structure of the printed version. It is a real drawback that there is no detailed electronic version yet available.

Although the Nature Encyclopedia of the Human Genome is a grand and pluralistic reference work - much of which will be of interest to even non-specialists - a synthesis of genomics itself remains an elusive goal.

ÁñÁ«ÊÓƵ

Csaba P¨¢l and E?rs Szathmary are members of the research group in theoretical biology and ecology, Eotvos University, Budapest, Hungary.