Published online 30 September 2010 | Nature | doi:10.1038/news.2010.500


Consortium solves its 1,000th protein structure

The international Structural Genomics Consortium celebrates its focus on expensive but exciting targets.

Structure 868, RECQL in complex with DNAThe Structural Genomics Consortium has focused on cracking the structures of human proteins — like the DNA helicase RECQL.Structural Genomics Consortium

An international team of 180 scientists, united in their goal to rapidly determine the structure of proteins related to human health, has solved its 1,000th structure.

It has taken the researchers of the Structural Genomics Consortium (SGC) six years and US$150 million to achieve.

Understanding a protein's three-dimensional structure is key to determining what that protein does. But unlike genome sequencing, in which the same technology can be used to sequence nearly any genome, solving the structure of one protein may require an entirely different approach to another. As a result, determining protein structures can be laborious and slow.

Enter 'structural genomics', an approach that focuses on speeding this process up. Several structural genomics projects began shortly after the draft human genome sequence was finished in 2000, but have drawn fire from researchers concerned that such large-scale projects siphon funds away from individual investigators.

"I just think this is an unwise way to be spending our money on science," says Gregory Petsko, a biochemist at Brandeis University in Waltham, Massachusetts. Petsko notes that at about $20 million a year, the SGC costs roughly as much as 100 National Institutes of Health grants to individual investigators. "Do you mean to tell me the science from these projects is worth 100 individual investigator grants?" he says. "I bet it's not worth 20."

The US-based Protein Structure Initiative, in particular, has been roundly criticized for tackling proteins of little interest to biologists. "'Structural genomics' has become a bad word," says SGC director Aled Edwards of the University of Toronto, Canada. "It's an albatross we bear."

But the SGC has always kept its eye on proteins with direct relevance to human health, and almost all of the 1,000 completed structures are human proteins (see slideshow). About 15% of the funding for the project comes from industry, with the rest from governments and charitable foundations such as the UK Wellcome Trust, according to Edwards. The list of protein targets is carved out in collaboration with SGC investors.

From structure to function

In its early days, the SGC aimed to drive down the cost of solving protein structures. But as the project progressed, the team realized that the focus on cost could short-change the science, says structural biologist Cheryl Arrowsmith of the University of Toronto and the Ontario Cancer Institute. Instead, the project changed tack and began to emphasize follow-up work that builds on structural information to develop a clearer picture of a protein's function. Edwards says that the team still tries to keep the price per structure below $150,000.


The result, says Arrowsmith, is papers such as last week's description of a small molecule that inhibits a protein called BRD4, which is involved in a rare form of cancer1. The authors of that study, including SGC researcher Stefan Knapp of the University of Oxford, UK, solved the structure of BRD4 bound to the inhibitor, and then went on to determine how the molecule interferes with BRD4 and whether it also slowed cancer growth. (It did, at least in cell cultures and human tumours transplanted into mouse models.) And best of all, boasts Edwards, the entire project took only seven months.

Results from the SGC are immediately published in public databases — often well before the team is ready to publish. "There is no first look, no six months delay," says Edwards. "This is a way of collaborating with industry that going to grow and become more common." 

  • References

    1. Filippakopoulos, P. et al. Nature advance online publication doi:10.1038/nature09504 (2010).


If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.

Comments on this thread are vetted after posting.

  • #60913

    This paper would have really helped me 5.5 years ago! I miss GPCRs

Commenting is now closed.