TRIESTE--The International Center for Genetic Engineering and Biotechnology (ICGEB) this year celebrates 10 years of serving biologists in the developing world. BioInform spoke recently with Sandor Pongor, head of the center's Protein Structure and Function/Biocomputing Group and node manager of ICGEBnet, about the anniversary.
Pongor explained that in 1990, ICGEB established the first login service for developing countrieshttp://www.icgeb.trieste.it. ICGEBnet was modeled after the European Molecular Biology net nodes (EMBnet) and soon joined EMBnet as a special node. Said Pongor, "For a long time our main activity was to provide the databases and GCG programs for the users that logged in through X.25 and the internet."
BioInform: How has ICGEB developed?
Pongor: Molecular biology cannot exist without computers, and many countries were in need of biocomputing at the time. We started modestly by using borrowed computer facilities and in 1990 we started our own computer resource, called ICGEBnet. Since then, manythings have changed. ICGEB now has 60 signatory countries worldwide and spends an estimated $300,000 on bioinformatics each year, including expenditures for fellowships, grants, a series of courses, and the maintenance of the computer resource, which is done by three people.
We have had over 700 students in computer training in Trieste. We also contribute to a biophysics PhD course and this year we organized a NATO workshop on structural biology and genomics http://www.icgeb.trieste.it/net/netcourse.html.
BioInform: What other bioinformatics resources are available to researchers in developing countries?
Pongor: When we started we were the only resource that accepted users from the developing world. In many cases, this was also the first scientific computer connection to the country. Networks have developed immensely since then and today there are many computer resources that can be used by anybody. But ICGEBnet is perhaps still the only one that provides free access and free hands-on training to the developing world.
EMBnet was of great help in the whole process, both as a cosponsor of our courses and by providing the lecturers. We joined EMBnet in order to keep our knowledge level with that of the European countries, and then to transfer it to the ICGEB member countries. This system seems to be working well now.
BioInform: Do you also conduct bioinformatics research at ICGEB?
Pongor: We are conducting research in two areas. The first concerns protein domain identification, where we developed the first publicly available domain sequence database, the SBASE domain library. The other subject area is DNA bending. We work on the hypothesis that bends are conformational signals in DNA that may have roles in such basic processes as gene regulation or transcription. Our most recent paper, which appeared in Trends in Biochemistry in September, will give an overview on how these interesting sites can be located in genome sequences.
Other than that, my research group of 10-15 members is interested in protein-DNA interactions and uses a variety of experimental techniques, including peptide synthesis, molecular cloning, and X-ray crystallography--not only computers.
BioInform: How many people use ICGEBnet?
Pongor: Over the years we have provided access to more than 1,500 users and have about 3,000 web-page contacts each month. With these figures we are in the middle range, as compared to big computer resources such as those in the UK.
In geographic coverage, we have users from 55 countries, and this is probably larger than what you see in other biological computer systems. I also think that our budget is probably the biggest single item spent on teaching bioinformatics in the developing world.
BioInform: Are there plans to expand?
Pongor: Bioinformatics is a rapidly changing scene. In the future the big resources such as the European Bioinformatics Institute in Cambridge, UK, or the National Center for Biotechnology Information in the US will have the primary role in providing access to genomic data, so the other computer centers will have to diversify. ICGEBnet will continue to concentrate on specialized services such as DNA bending and protein domain searching. Naturally we also want to continue good basic coverage and to develop ou courses. We want to continue in our next 10 years as a useful service in bioinformatics.