From: Harvey Newstrom (mail@HarveyNewstrom.com)
Date: Tue May 13 2003 - 16:55:05 MDT
gts wrote,
> Brian Atkins wrote:
>
> > There is a chart in the latest LEF mag that shows Metformin as
> > producing an almost 100% match in gene response as compared to CR
> > when studied using gene chips.
>
> Hmm, looks to me like that bar-chart is very misleading. They set the
> metformin bar at 100% of "maximum effect at reproducing caloric
> restriction"
> for the drugs they studied. The chart thus gives the visual
> impression that
> metformin mimics 100% of the gene expression changes from CR. But in fact
> that is not what the chart actually indicates.
I agree! I noticed this myself when I read the article. They "set" to
metaformin bar at 100% and measured other substances against it. This is a
common technique when the theoretical maximum range is not known. However,
it does produce the misleading visual that best result thus far is the final
solution, because it appears to be "100%" effective (measured against
itself). How could anything else be discovered that is even more than 100%
effective?
-- Harvey Newstrom, CISSP, IAM, GSEC, IBMCP <www.HarveyNewstrom.com> <www.Newstaff.com>
This archive was generated by hypermail 2.1.5 : Tue May 13 2003 - 17:08:42 MDT