Nov. 25, 2003 — Data, data everywhere. But not the data you need. That’s how some scientists working with bioinformatics can feel. The sheer amount of gene and protein data boggles even their minds.
null
That’s where TEMBLOR comes in. It’s a $23 million European Union-supported program to build and manage several databases for Europe’s biomolecular researchers.
 |
null
“There has been much research on the genetic sequence of human beings and proteins,” said Frederick Markus, project officer for TEMBLOR at the European Union. “So there is a huge range of data and experimentation out there, but you need to put it in databases with computer software and analysis to make sense of it.”
null
To get an idea of the sheer amount of information to compute, consider this: There are 30,000 genes in the human body, but exponentially more proteins. And then there are the other species — plants and animals — that have been sequenced. If researchers are to move ahead, they need quick and easy access to data from experiments performed by their peers.
null
The European Bioinformatics Institute (EBI) in Cambridge, England, has managed TEMBLOR since its launch in early 2002. The state-funded center specializes in biological database management, including nucleic acid, protein sequences and macromolecular structures.
null
“EBI is without a doubt the biggest bioinformatics supplier in Europe,” said Graham Cameron, EBI’s associate director. “We get half a million hits on our site a day.”
null
In the nearly two years since its launch, TEMBLOR has built several very specific databases, including DESPRAD, for gene expression experiments and IntAct, for protein-protein interaction data. Another, called Integr8, acts as a telephone exchange to connect researchers to the other databases.
null
The goal is to give scientists a simple interface while carrying out complex queries across databases.
null
Cross-referencing is essential because scientists often use studies of other species to gain insights into human genetics.
null
“Take the yeast cell, for example, 40 percent of its genetic structure is the same as humans,” Markus said. “It’s surprising to hear that people contributing to cancer research are working on yeast cells, but you can learn a lot about the cell cycle by studying yeast.”
null
Those involved in TEMBLOR say the most challenging aspect of the project has been the need to standardize all the data.
null
“Gene expression is very sensitive to experimental conditions,” EBI’s Cameron said. “You really need to record all the detail of the experimentation” and do that in a standardized fashion. To achieve this, scientists have been asked provide their data in a specific format, so standard software can harvest the information.
null
It’s estimated that the TEMBLOR databases serve a community of 100,000 biomolecular scientists around the world. Much of the time, users are also contributors.
null
Laurent Duret is a French scientist who specializes in evolutionary biology and bioinformatics. He is the TEMBLOR representative at the University of Lyon 1 in central France. His lab at the Lyon Bioinformatics Center is one of many in Europe building the TEMBLOR databases. In exchange, the lab receives about $300,000 over three years.
null
“Our job is to provide the maximum amount of information on various genomes,” Duret said. “We are providing data that will allow comparative analysis of all the genes that have been sequenced in the human genome.” That means collecting and inputting data for the 117 species that have been sequenced.
null
Some of those involved in TEMBLOR fear European Union funding might not be renewed when the project finishes at the end of 2004.
null
“It would be a great loss to the scientific community,” said Alwyn Jones, a structural biology professor at Sweden’s Uppsala University and another TEMBLOR contributor.
null
For Cameron at EBI, the standardization that TEMBLOR has imposed has been pivotal.
null
“Normally what happens in a new area of science is that it starts out in shambles,” he said. “The achievement here has been the building of (scientific) community consensus. Never has there been an instance where the scientific community has marshaled itself like this in terms of standardization.”