The two scientific teams who jointly announced the completion of the first draft of the entire human genetic code last June will publish their results this week. The international consortium's version will appear in Nature this Thursday, while that of US Company Celera Genomics will be published in rival journal Science on the following day.

Both groups report that the total number of human genes appears to be much smaller than previous estimates of between 50,000 and 100,000. The Science paper, by Celera president Dr Craig Venter and 282 co-authors, estimates that the human genome contains between 26,000 and 39,000 genes. The publicly-funded Human Genome Project's (HGP) article in Nature puts the figure at between 30,000 and 40,000.

The results have surprised those who assumed that humans would have many more genes than simple life-forms such as the nematode worm, which has around 18,000 genes. But the human genome is more versatile than that of a worm - many human genes are composed of 'subunits' that can be reshuffled to make a variety of different proteins. Professor John Sulston, former head of the UK's genome sequencing effort, also points out that humans have many more 'control genes' than worms, which in turn can influence other genes in a variety of ways.

The publications have reignited the disputes over data access between the two groups. Sulston accused Celera of underplaying their reliance on the HGP's data to complete their own map. But Venter dismissed the claims as 'sour grapes', saying his version was superior to the HGP's freely-available one.