Sir

Public debate about advances in molecular biology and the billion-dollar revenues that may accrue from new therapies has taken on particular significance in Brazil. This is a result of the enormous success of the consortium which reported the genetic sequencing of Xylella fastidiosa1, the bacterium causing the 'amarelinho' disease that destroys more than 30% of São Paulo's orange groves.

The same consortium is making great sequencing progress in three other areas: human cancer genes endemic to Brazil; genes related to sugar-cane metabolism; and the genome of the Xanthomonas citri bacterium which causes citric cancer, widespread in Brazil's orange groves.

Brazilian discoveries need to be protected, but the country's legislation does not permit the patenting of live beings. Brazilians are inexperienced in dealing with international patents. Of the 100,000 or so patents granted by the United States patenting office each year, a mere few dozen go to Brazilians.

Every original invention that is useful and has a commercial potential can be patented somewhere in the world. Human imagination knows no bounds: take, for example, US patent 5443036, a device for encouraging a cat to exercise by chasing a light spot2. However, the US statutes define four types of inventions for purposes of registration: new processes (or methods); machines (or devices); manufactured articles; and new compositions of matter.

Processes or methods are inventions that describe how to do something. A typical claim could be “a method to make vegetable soup”, which comprises all the necessary steps for preparation. Devices are machines that do something, such as “brush teeth automatically”, followed by a description of the parts that make them up and how they interrelate. An example in the third category might be an “optical fibre”, with a detailed description of its structure and composition.

In the fourth category, although matter existing spontaneously in nature is not patentable, new compounds and chemical compositions can be. Thus, a synthesized bioactive glass for the substitution of bones and teeth, containing oxygen, silicon, sodium, calcium and phosphorus, has been patented3. But its constituent elements, being natural, cannot. (One can imagine the possible consequences had each element of the periodic table been patented following discovery.)

A patent requires human intervention in the design, construction and synthesis or manufacture of the product. Another requirement is descriptive sufficiency: enough information for the invention to be reproduced. Hence there is a clear difference between invention and discovery. The former results in a new composition, product, device or process; the latter from the unveiling of universal laws or from the structure or composition of extant natural matter.

Powerful lobbies have encouraged the filing of patents on live beings. North American and European agencies — with the exception of the French — have granted several thousand patents for genes and genomes, despite the promises of world leaders. The most common argument is that patents will be granted only after the gene's functionality has been clearly established. But surely this is a question of discovery, not invention?

From an ethical standpoint, genes should not be patented (although medical inventions based on such a discovery can and should be). Yet in industrialized countries researchers are filing thousands of gene patents: Celera Genomics alludes to 6,500. So how should Brazil, or any country, protect the public resources that make these discoveries possible? Until logic and common sense prevail, we should patent our inventions everywhere. However, one could deposit discoveries that might have immediate commercial interest in electronic databases and charge for privileged access. This would not prevent other investigators continuing their research, even if they could not afford privileged access to the latest discoveries.