[This is a sidebar to "The State of Mutation Curation". Click here to return to main story.]
By Allison Proffitt
March 29, 2011 | “It’s important to realize that each gene—each 20,000 of them—needs a person to look after it properly, in a way that’s accurate. And the labs around the world all need to put their data in. It’s got to be a distributed system... It’s critical to divide the load.”
Richard Cotton believes that the Human Variome Project (HVP) will help manage this distributed labor. Cotton is scientific director of the HVP and head of the Genomic Disorders Research Centre in Melbourne, Australia. He has seen the HVP mature from the Human Genome Organization’s database initiative to the Human Variome Society to an incorporated nonprofit body in May 2010.
The vision of the HVP is to facilitate the establishment and management of standards, systems, and infrastructure for the “worldwide collection and sharing of all genetic variations effecting human disease,” with a “Golden Goal” of facilitating the sharing of data on over a million cases of genetic disease by the end of 2015. “The HVP is a consortium of individuals and research institutions that are all doing work in this field and working toward a single vision,” says Timothy Smith, communications officer at the HVP.
The HVP will collect data in country-specific nodes as well as in gene- or locus-specific databases. “The reason we’ve got two [avenues] is so that we capture everything,” says Smith. “It’s a way of ensuring complete coverage.”
HVP organizers realized early on that because of differing ethics regulations in various countries, “producing some sort of centrally-mandated solution or one-size-fits-all approach was going to be almost impossible.” Hence, HVP country nodes are built and resourced by individual countries. There are currently seven HVP nodes: China, Greece, Belgium, Egypt, Kuwait, Australia and Malaysia. (Cotton’s Genomic Disorders Research Centre operates the Australian node along with other organizations.)
Meanwhile, disease- or locus-specific databases gather data on one indication. Cotton stresses that the community will be important for these databases, highlighting, for example, the role of the colon cancer community in “loading up their database in a big way.” Cotton adds: “We don’t have our own databases, but we’re working to encourage people to collect the stuff into these locus-dependant databases.”
Ultimately, HVP hopes that all genetic data will be collected according to HVP standards and best practices, expertly curated, and then deposited into central databases such as OMIM, the UCSC Genome Browser, and those at the National Centre for Biotechnology Information (NCBI) and the European Bioinformatics Institute (EBI) and made freely available.
But free data costs money. “There are many people around the world applying for funding with and without our assistance, and they’re all having trouble,” Cotton says. “The governments just don’t seem to see it as an issue. I believe it’s because inherited disease is rare and each one is rare and doesn’t have a voice.”
The HVP did announce a $300-million investment from China in January, but Cotton and Smith stress that the money isn’t lining HVP coffers in Melbourne. This funding “is not coming to us here in Australia,” said Smith. “It’s being spent in China by the Chinese to contribute toward projects that will further the goals of HVP.”
China will set up their country node, and intends to also set up 5,000 to 8,000 locus-specific databases. Smith suggests that some of the Chinese money could be used to support existing gene databases, perhaps where there are funding difficulties, rather than just creating new ones. Cotton reckons China should consider funding “a key group of curators to make sure that the data’s collected and once it’s collected, it’s really good.”
Making sure that variome data is collected and curated is a monumental challenge. “There’s the legacy data that’s in people’s records now, and there’s all the stuff that’s coming off the sequencers. The stuff that’s in people’s labs and records is going to be a bit more time consuming to get, whereas if we can get the informatics guy to take the stuff off the sequencer and put it into the database, that’s much cheaper,” Cotton says.
HVP does not want to get hung up on platforms people may use to build databases, says Smith, listing the MutaDATABASE and others. “We’re in touch with the developers of all of these software platforms, working with them to make sure that the software they’re producing is useful to the community.”
HVP’s prime concern is that the data are available and efforts are not duplicated across different databases. “There really only needs to be one and people that are interested in studying these databases should work together to make sure that there is one authority,” Smith says.
HVP doesn’t see other databases as competitors. “There are pros and cons of all of them,” says Cotton, “and we hope to work together in the end.” The key is to divide the load. “If you split it up, you can do it and people are doing it if they get the right funding!” says Cotton. “But what we’re trying to do is make sure that everything is available.”