Banner
    Is Science Too Technology-Dependent? Are 'Artifactual' Findings Up?
    By Greg Critser | August 9th 2012 02:11 PM | 5 comments | Print | E-mail | Track Comments
    About Greg

    Greg Critser is a longtime science and medical journalist whose work appears in the LA Times, the Times of London and the New York Times. He is...

    View Greg's Profile
    Excuse my use of the personal pronoun in this short article, but I am writing to get input as much as I am to simply observe a phenomenon.

    I am increasingly drawn to the connection between bench science and the companies that provide them with their tools - from commercial assays, probes, and imaging chambers to engineered animals and so-called "experiment in a box" products.

    As a journalist I am drawn to the subject for many reasons: it is an industry that is almost completely uncovered and one that supports, via advertisements and sponsorships, a growing number of science magazines and science venues. 

    And there seem to be a growing number of 'artifactual' findings due to reliance on them.

    Perhaps the best known in recent years is the case of resveratrol co-discoverer David Sinclair's trouble with a SIRT probe.  Sinclair's SIRT compounds earned a $750 million dollar acquisition price tag (from Glaxo) based on their ability to "cleanly" target one SIRT gene. That's what made it such a desirable acquisition for drug development. 

    As it turned out, the experiment that supported that claim was deeply flawed. The results turned out to be artifacts of a commercially-available florescent probe.  It is unclear whether or not this artifact had something to do with SIRT's failure (so far)  as a diabetes drug, but it certainly did not buttress Glaxo shareholders' faith in their giant investment.  And, surprise, Sinclair's jealous detractors had a schadenfreude bacchanal with the whole thing, replete with unsubstantiated party-talk about fraud and misconduct. 


    Credit: Shutterstock

    Commenting on the modern lab's need for such tools, scholar Mark Willingham of the Department of Pathology at Wake Forest noted in Science: " 'Companies are taking advantage of known pathways for apoptosis and developing specific regulators.'  Scientists who must cope with apoptosis in broader investigations, rather than trying to understand it, have different requirements. 'They need user-friendly kits that don't take time to learn...For them the name of the game is convenience.' "

    Concern about over-reliance on packaged experimental tools has, so far, come from older investigators who remember the days when researchers had to develop their own tools. As the onetime dean of mammalian physiology,  University of Texas's Ed Masaro, once noted, "reliance on such technology is making it possible for young researchers to render findings about results in mammals but without understanding the basic mammalian physiology behind it."  And so troubled by the trend was one leading investigator that he "told him to get out until he could figure out something on his own."

    I'd like to know your experience or thoughts on this:

    Have such tools made 'artifactual' findings more likely? 
    Can you cite any examples?
    What's the price of reliance?

    If you can't comment publicly, please write me through my profile page.

    Comments

    Gerhard Adam
    Is this really any different than someone purchasing software that claims to calculate some phenomenon or other, or to model or simulate some behavior?  Without understanding the algorithms involved and establishing the difficult bits about such software, then the individual utilizing it is simply relying on blind faith that the authors knew what they were doing.

    'They need user-friendly kits that don't take time to learn...For them the name of the game is convenience.'
    While I can appreciate this sentiment, to a degree, it reeks of "productivity" and not science.  If it isn't worth learning, then it probably isn't worth using.
    Thanks. Please tell me if you know any instances in the life sciences that I should look at re artifacts that are consequence of commercial bench assays etc

    Gerhard Adam
    I don't know if this is the kind of thing you had in mind, or if you've already seen stuff like this.
    With its engaging and intuitive design, iReport for Real-Time PCR can be used by bench scientists to easily and accurately understand the biological meaning of their experiment within minutes.
    http://www.ingenuity.com/news/html/pr_120402_life_technologies_ingenuity.html
    randallmayes
    Greg- a timely example is testing Olympic athletes. When testing for EPO, the first assay is from urine. If the test shows high levels, it can be a false positive or the result of altitude training. The doping agency may perform test B, from the blood, to determine if there is a marker for synthetic EPO.
    Randall Mayes
    @ Randell
    I didn't know that altitude training could show high levels in the results, thats really interesting. Makes me wonder why most of the athletes that are taking the EPO don't just stay they were training in the mountains to get away with it. I think its just sickening in the first place how these athletes are taking drugs to do better, whats the point? Might as well have a bunch of androids and robots join in the Olympics.