the United States to Europe to receive theirlife-saving drug. What a great perspectivefor Europe’s biotechnology and economy!Hartmut MichelMax Planck Institut of Biophysics, Heinrich-Hoffmann-Straße 7, 60528 Frankfurt am Main, Germany

1. Schiermeier, Q. Nature 406, 111 (2000).

2. Smaglik, P. Nature 404, 322 (2000).

3. Cormier, E. G. et al. Proc. Natl Acad. Sci. USA 97, 5762–5767

(2000).

No room on the carouselfor meeting of like minds Sir — The day has finally arrived when Ireceive more scientific meetingannouncements than junk mail. I read allthe programmes with interest, but wonderabout the rationale for such a deluge, wheninformation flows freely over the Internet.Notices of meetings and workshops,scientific discussions, the contents offorthcoming journal issues and many othermatters can all be found on the Web. So whytravel to a meeting when, in most cases, theinformation is not even new?

As I open the envelopes, my interest fadesrapidly when I see the same speakers andtopics regularly repeated. I am not criticiz-ing scientists who circle the globe sharingtheir discoveries and knowledge with therest of us, but to me they are on a scientificcarousel. Each colourful horse is ridden by aspeaker with a lecture in his or her righthand. As the merry-go-round rotates, thetopic and speaker will cross your path nomatter where you are standing.

So what can justify the existence of scien-tific meetings? They ought to allow us tocommunicate and share new advances; todiscuss new applications for a technique; tohear the latest evidence supporting a discov-ery; to develop new projects, and so on.

But the whole principle of a scientificmeeting — innovation — is ignored in mostof these scientific events. The number ofmeetings inversely correlates with theamount of interesting data to be communi-cated. On the scientific carousel, a meeting isheld for commercial reasons and peopleattend out of opportunism (local rivalries,for example) or to visit fancy places. Inmeetings held by scientific societies, how-ever, the spirit of science is maintained, asparticipants share and discuss advances andchallenge each other verbally.

An individual can communicate excitingresults on the Web, but the work cannot begenerally accepted until it has been chal-lenged and recognized by fellow researchersat meetings. The recognition and respect ofone’s peers is what constitutes scientific kinship and justifies society meetings. As

correspondence

NATURE | VOL 407 | 21 SEPTEMBER 2000 | www.nature.com 285

for the scientific carousel — like any otherfairground attraction, time and marketforces will determine its lifespan. Eleuterio R. Hernandez Instituto de Bioquimica (CSIC-UCM), Facultad deFarmacia, Universidad Complutense, 28040- Madrid, Spain

Standardized addresseswould make web easier Sir — This may be the age of information,but gaining access to it is often difficult.Finding a site on the World-Wide Web canbe like searching for a needle in a haystack.

In India, only half of the 40 laboratoriesin the Council for Scientific and IndustrialResearch (CSIR), the country’s largestresearch and development organization,have websites of their own. These 20 institu-tions follow five different ways of namingtheir sites. If you don’t know the address, youcan’t reach the site by guessing. None ofthem end with .gov.in. The Central Mechan-ical Engineering Research Institute, Durga-pur, has opted for www.cmeri. com. Someend with .org — which gives the impressionthat the pages are hosted by a serviceprovider located in the United States, whereweb addresses do not end in a domainindicative of the country. Most web-basedsearch engines are unable to help, since theyonly reach some 16% of the web (see Nature400, 107–109; 1999).

Adopting uniformity in naming domainscould circumvent this problem. Many insti-tutions around the world use their name oracronym as the lowest domain and move uphierarchically. For example, all the institu-tions in the CSIR could use .csir.dsir.gov.in,since CSIR is part of the Ministry of Scienceand Technology’s Department of Scientificand Industrial Research. The Centre for Cel-lular and Molecular Biology could be atwww.ccmb.csir.dsir.gov.in. Educationalinstitutions should follow the same conven-tion. By understanding the organizationalhierarchy of any institution one should beable to guess the web address.

The next improvements could be to thecontent, organization and facilities providedby the web pages. In the scientific world,concerted efforts are under way to realise thefull potential of electronic communication,with e-print archives, online journals andthe development of E-Biomed. Web pagesfor government-funded research and devel-opment laboratories need to offer more thanjust basic facts about the organization. Theyshould show depth of content and shouldinclude details of their scientific activities,search facilities and routine updates.Sasidharan RajkumarNagarjuna Group, Nagarjuna Hills, Hyderabad 500082, India

Allowing gene patentscould be an expensivemistake for the US Sir — In the discussion of gene patents1, it isgenerally overlooked that these primarilyaim at the encoded protein and not the gene.The discoverers try to patent the use ofthe ‘gene product’ either as a therapeuticprotein, or as an intervention point (‘drugtarget’) to treat a disease.

However, although the gene sequenceprovides necessary information on the pro-tein structure, this information is insuffi-cient, because the mRNA and the originalgene product, the protein, are modified in anunpredictable way depending on the tempo-ral and spatial context of their biosynthesis.Who could have predicted the active form ofinsulin from its gene? Also, the co-receptorof HIV, CCR5, on which a controversialpatent has been issued2, appears to be a co-receptor for HIV only if it is modified byattachment of sulphate residues3. It is evi-dent that, if a patent covers a ‘gene product’,it should only give rights on the predicted,unmodified form. Experimental evidencefor the existence of the predicted function ofthe gene product should be provided.

What about patenting gene products asdrug targets? The EU patent directive statesclearly that the human body and its compo-nents in their native environment cannot becovered by a patent. What is a drug targetother than a component of the human bodyin its native environment? This statementcan only mean that drug targets are notpatentable in Europe, otherwise it would benothing but a political declamation.

There are other problems with patentingdrug targets. Many drugs are still discoveredusing animal model systems or cellularassays. In the United States, for example, theNational Institutes of Health has a numberof screening programmes using mice todetect anti-epileptic compounds, and cellcultures to detect cytotoxic compounds,where the molecular target is unknown. If ascientist or a company discovers a drugusing such an assay, they would be stupid totry to identify the drug target, because thereis some likelihood that the drug target isalready covered by a patent. Patenting drugtargets is therefore scientifically counter-productive: it enforces a culture of ‘notwanting to know’.

Let’s assume that drug targets can bepatented in the United States but not inEurope. As a consequence drugs could bedeveloped unrestrictedly in Europe but notin the United States. US companies wouldthen be forced to move their drug discoveryand development to Europe. Still thesedrugs could not be used in the United States.As a result patients would have to travel from

© 2000 Macmillan Magazines Ltd