Gene Wikiality


0 Flares



Filament.io





Made with Flare More Info‘>

0 Flares


×

Two years ago Stephen Colbert, host of the news-parody show, “The Colbert Report” coined the word “wikiality” to describe a reality defined by the majority.

“Nation, it’s time we used the power of our numbers for a real internet revolution,” Colbert told his audience. “Together we can create a reality that we can all agree on — the reality we just agreed on.”

Like Colbert, genetics researchers also hope Internet users will help realize their wiki vision — an online database with material sourced from the majority.

Earlier this month, Jon Huss III and his colleagues from San Diego State University, the Genomics Institute of the Novartis Research Foundation and Washington University School of Medicine outlined their proposal to develop a gene wiki on Wikipedia.

In a PLoS Biology paper, the researchers described their plan to create an online resource filled with content that can be read by all audiences, from lay people to students to professionals. And unlike extant gene databases such as Entrez Gene in which large tracts of information are submitted and edited by a few experts, they said, the entries in this database would come from a projected vast pool of volunteer contributors.

(On a side note, the gene wiki isn’t alone in looking for talented contributors of information about genes. 23andMe is looking for a full-time scientific curator — someone who can, among other things, sift through journal papers and help identify genetic associations of interest. For further information, check out the job posting here.)

Huss and his colleagues aren’t the first to think of a gene wiki, but they are unusual in going against the top-down approach by laying a foundation on Wikipedia. By the time their paper was published, they’d created entry stubs for 7,500 genes such as ITK, and updated entries for another 650, including BRCA1.

“Because articles are dynamic and not subject to rigorous peer review, the gene wiki is not intended to be a reference that is cited in a traditional peer-reviewed article or used exclusively as a source of gene annotation,” the researchers wrote.

Huss and his colleagues hope to involve a number of people, each taking a little time to enter and edit these gene entries on Wikipedia. One obvious question is how long it’ll take people unfamiliar with the site to learn how to contribute. Early feedback suggests the process may not be as intuitive as expected, and notes that participation is key for this effort to succeed.

But the researchers’ plan to develop a gene database that can be accessed and edited by anyone at any time also brings up the same issue that has dogged Wikipedia over the years: accuracy.

Wikipedia’s everyman-as-editor policy has led to some high-profile controversies such as the Siegenthaler incident, in which a retired journalist was falsely described on Wikipedia as being involved in the Kennedy assassinations. There have also been cases of political staffers editing the entries for their bosses — and their bosses’ opponents.

Stephen Colbert has played on Wikipedia’s multiple-editor feature as well, asking audiences to log in and change entries involving elephants, George Washington’s slaves and the true meaning of Warren G. Harding’s middle name.

Wikipedia co-founder Jimmy Wales, who is delighted with the researchers’ plans to have the site host the gene wiki, has acknowledged that the online encyclopedia isn’t perfect.

“[Wikipedia] should be thought of as a work in progress — it’s our intention to be Britannica or better quality, and our policies and everything are designed with that goal in mind. We don’t reach that quality yet — we know that,” Wales said in December 2005.

Surprisingly, Wikipedia’s perceived Achilles heel is what also seems to be helping the online encyclopedia attain Wales’ stated goal. In allowing anyone to make changes at any time, the majority’s decision has final say, and it turns out that the majority favors accuracy.

A study conducted by the journal Nature compared the number of errors found in matching science entries from Wikipedia and the Encyclopedia Britannica, long regarded as the gold standard of references. The average Wikipedia entry had four errors; the average Britannica entry had three.

Wikiality, it seems, may not be that different from reality after all.

Still, for the gene wiki to become what the researchers envision, they’ll need informed people — lots of them — who are willing to log in during a coffee break or three, check out an entry or two, and make necessary edits and additions. They’ve built it; it’s time to see if the scientific community will come.

Image from: Huss et al, PLoS Biology 2008.






0 Flares



Twitter


0







Facebook


0








Google+


0







Email


0








LinkedIn


0








Filament.io





Made with Flare More Info‘>

0 Flares


×


  • ejain

    Does “the majority’s decision” really have final say on Wikipedia, or is it the site admins who do? In fact one issue some scientists have with Wikipedia appears to be that their real-world authority doesn’t translate into authority in Wikipedia. I believe Citizendium tries to address this to some extent.

    Piggy-backing on Wikipedia seems better than setting up yet another Wikipedia-like system that doesn’t add much in terms of functionality and is moreover less likely to be as durable. I’m kind of surprised that the Wikipedia people don’t object to such automated seeding, but apparently they don’t.

    Would of course be nice to have wiki-like sites that are more suitable for sharing scientific data (better handling of data provenance and conflicting opinions etc), but talk is cheap…

  • sjors

    The thing about Wikipedia’s reliability that most people don’t seem to understand yet, is that it remembers *everything*. You just have to look beyond the most recent version of the page. If you have any reason to doubt the content, just look at older versions and at the discussion page. That is a very powerful indication of reliability, but it takes some skill to interpret.

    Of course, better tools are needed to process this type of information automatically. Hopefully these will include as much information about the authors, e.g. their other contributions (this is already being done) and their social network. Ultimately these tools should also look at the actual meaning of their changes and see the difference between fixing a spelling error or a defamatory comment.

Return to top
0 Flares Twitter 0 Facebook 0 Google+ 0 Email 0 LinkedIn 0 Filament.io 0 Flares ×